×

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

Bi-monthly release with minor bug fixes and improvements

Getting started with KStars EKOS and INDI

  • Posts: 271
  • Thank you received: 72
Hi.

First post here so let me know if i should add more[less] info, or if i'm off topic..

Last week i started installing my remote observatory. At the moment i am only testing software before i start building the "box obsrevatory"

I have been reading up and have now kstars-bleeding on my workstation. I also have compiled Astrometry.net for the workstation.
On the server side i have compiled INDi from source + 3rddparty/indi_qhy + indi_gphoto. I also have gsc on the server.
The instructions i have followed for kstars-bleeding, astrometry.net and INDI + 3rdparty is from this thread www.indilib.org/forum/general/210-howto-...st-libindi-ekos.html

The issues i'm having is:
1. Autoguiding only works when i'm selecting Guider : CCD Simulator Guider, Via: Telescope Simulator. (in ekos)
I have tried getting guiding to work with my QHY5 (Orion SSAG) selecting Guider: CCD Simulator, Via: QHY5.. (in ekos)
When i do this everything seams to work fine, the stars moves when drifting in RA and DEC, but i get a message: GUIDE_RA failed to reach start after 30 iterations.
I can see that after ~25% of calibration the star only moves right in a sig-sag pattern, but newer left. Looks like the guider can only move RA in one direction.
Am i using correct settings to test / simulate the QHY5 and Celestron AVX for guiding? Still testing inside, so have to simulate the imaging for guiding.

2. In INDI control panel Alt/Az have 30deg misalignment. If i move to 1/181 Alt/Az in ekos (slightly above the horizon due south), then INDI panel shows -29/181.
So if i want to park at that point i have to set park position -29/181. This works fine until i unpark. The recticle for the mount will move to a altitude of -29. If i zoom in on the recticle before i unpark, and then do a sync after unparking everything works. If i forget to do the sync, i can end up rotating the dec motor 360deg, possibly creating a crash situation.

3. Minor.. Had to download new stuff manually and find the correct directory for the content. Loading of providers from file: edu.kde.org.... failed.

Only been installing and testing for a few days, and am weary impressed so far. No doubt it will result in a working obsy :cheer:

Some specs:
Mount Celestron AVX,
Guider QHY5 (Orion SSAG),
CCD Canon eos 550d (rebel t2i)

Workstation:
Linux 3.19.0-28-generic x86_64 GNU/Linux (Ubuntu)
15.04 (Vivid Vervet)
kstars 2.4.0
Using KDE Frameworks 5.9.0
This program is part of the Astrometry.net suite.
Revision 0.61-4-gc3ce036, date Fri_Sep_11_10:35:20_2015_-0400.

Server:
Linux 3.19.0-28-generic x86_64 GNU/Linux (Kubuntu)
15.04 (Vivid Vervet)
2015-09-12T13:50:57: startup: indiserver
INDI Library: 1.1.0
Code $Rev$. Protocol 1.7.

Not sure what logs i could add for this request. Please request some if they can help.
Hope i'm not to far f topic, and that someone has some tips :)
Proud owner of Observatory 17b - A remote Linux observatory.
Website: Observatory 17b
Build thread @ SGL: Starting summers observatory project
8 years 6 months ago #5042

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

Hello! Glad you find INDI/Ekos useful!

1. QHY has problems guiding. In fact, I made a test this morning, and then I emailed QHY and made this post . I already got an email from QHY saying they're looking into it.
For me, the only direction that doesn't work is SOUTH. Can you try doing a dry run? Connect QHY guide cable to your mount; Go to QHY in the INDI Control Panel, and select the "Guider" tab. Enter manually the guiding pulses one by one (1000ms is OK). Then try each guide pulse, and then check the celestron RA/DEC. Make sure that if you set NORTH to 1000, SOUTH is set to 0 and vice versa. I got all pulses resulting in motion except in SOUTH. Do you get the same behavior? Report RA/DEC after each pulse.

2. It's not suppose to be negative if it was really above the horizon!! But I wrote the driver without testing it on a celestron mount as I don't have one, so I guess you can test it then. So next time before you connect, go to INDI control panel, Celestron tab, then select Options tab, then enable debug, and then below enable verbose and scope logging. Then select also "Log to file". It will generate a complete log file. Then go again to 1/181 and after it reports -29/181, just send me the log file. And more importantly, tell me what the Celestron handset reports the current Alt/Az as? is it 1/181 or something else?

By the way, it's you're using Ubuntu, you don't need to compile anything, it's all in the PPA.

On the workstation, you can simply do:
sudo apt-get install astrometry.net
sudo apt-get install kstars-bleeding kstars-bleeding-dbg

On the server side:
sudo apt-get install libindi-full indi-dbg

That will install libindi and all 3rd party drivers. If you JUST want QHY + GPhoto with libindi
sudo apt-get install libindi1 indi-dbg indi-qhy indi-gphoto

And you're set!
Last edit: 8 years 6 months ago by Jasem Mutlaq.
8 years 6 months ago #5043

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

  • Posts: 271
  • Thank you received: 72
hi.
I went trough the steps you explained.

For the QHY (I did mention i'm using the Orion Star Shot Autoguider with the indi_qhy_ccd driver?)
Start: 21h 08m 43s -28deg 01' 53"
North: 21h 08m 42s -28deg 02' 13"
South: 21h 08m 42s -28deg 01' 53"
West: 21h 08m 40s -28deg 01' 53"
East: 21h 08m 40s -28deg 01'53"

So on my end looks like East does nothing, and West not doing much.

For the Alt Az.
Starting from a initial position of the mount. Marks aligning pointing towards polaris.
The INDI driver reports 90:00:00 / 90:17:48 Alt / Az. Should a Alt / Az mount be pointing straight up when starting up, but a EQ mount not?
Slew to -26 / 181 in INDI Control Panel, EKOS "and KStars" reports 1deg 59' 40" / 182deg 27' 04" Alt / Az

Attaching images and log from celestron_gps session. Mount is Celestron AVX, Hand controller is Celestron NexStar +





File Attachment:

File Name: 2015-09-12.txt
File Size:35 KB
Proud owner of Observatory 17b - A remote Linux observatory.
Website: Observatory 17b
Build thread @ SGL: Starting summers observatory project
Last edit: 8 years 6 months ago by Magnus.
8 years 6 months ago #5044
Attachments:

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

Thanks, that's very helpful. Few more things:

1. What's your Latitude/Longitude?
2. Are Ekos Mount tab values for Az/Alt the correct one? If you're looking north at Polaris, your Azimuth is close to zero and you're Altitude is equal to your site Latitude or very close to it.
3. Is the RA/DEC in INDI control panel correct and agrees with the Celestron handset? Just to make sure here as well.
Last edit: 8 years 6 months ago by Jasem Mutlaq.
8 years 6 months ago #5045

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

  • Posts: 271
  • Thank you received: 72
For testing i have set the time-site for Oslo Norway, about 59N / 10E.

The ekos alt / az is ~61 / ~0 when mount is at index, pointing +- polaris, so should be ok.

Indi and ekos ra / dec is equal and corresponds to celestron hand controller.

Looks to me that it's only indi alt / az that is showing wrong values. I also notized that when ekos was showing alt / az ~61 / ~0 indi was at alitude 90deg00'000". Looks like it assumes a fixed value? See first image.
Proud owner of Observatory 17b - A remote Linux observatory.
Website: Observatory 17b
Build thread @ SGL: Starting summers observatory project
Last edit: 8 years 6 months ago by Magnus.
8 years 6 months ago #5046

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

From Celestron's manual, it seems that both Az and Alt are set to 0 once alignment is successful. It seems 'zero' is the value 0x4000 out of a maximum value of 65536 (unsigned short maximum). At any rate, I threw in some math magic and it appears to work in the simulator at least. So please try again with the logging turned on again and let me know how it goes.
8 years 6 months ago #5047

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

  • Posts: 271
  • Thank you received: 72
Hi.

This time i powered on the mount, started indiserver, kstars and ekos, and then did the quick align in NexStar to allow some more logging.

After Celestron GPS: Debug is enabled RA-DEC is (12:00:00, 0:00:00) and AZM-ALT is ( 90:00:00, 90:00:00)

Found a firmware version nr in the NexStar if it can be of help.
HC:GEM 5.22.3141
MC:7.08.3071

I also double checked that indi, ekos, kstars and nexstar is showing same RA DEC

File Attachment:

File Name: 2015-09-13.txt
File Size:86 KB
Proud owner of Observatory 17b - A remote Linux observatory.
Website: Observatory 17b
Build thread @ SGL: Starting summers observatory project
8 years 6 months ago #5051
Attachments:

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

Sorry forgot to say you need to compile it or wait until tomorrow PPA update.
8 years 6 months ago #5052

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

  • Posts: 271
  • Thank you received: 72
Should i rebuild just INDI on the server side, or Kstars - Ekos also?

I have rebuild indi, and now alt / az is 59 / 0 at start, but still slews to same location.

In Ekos ~ 28 / 181 same as before
In Indi Control -29 / 91..




File Attachment:

File Name: 2015-09-13-2.txt
File Size:36 KB
Proud owner of Observatory 17b - A remote Linux observatory.
Website: Observatory 17b
Build thread @ SGL: Starting summers observatory project
8 years 6 months ago #5053
Attachments:

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

You only need to do it on the server. Sorry, I updated it again, I forgot to change the slewAzAlt function. Please try again with the logs.
8 years 6 months ago #5054

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

  • Posts: 271
  • Thank you received: 72
Updated to new code.

Not sure what is happening now. Almost seams like dec and ra is inverted? Not 100% sure.

Log is showing slewing to 1 / 181 by clicking on position in kstars and selecting slew.
This slews the scope to correct position, showing 1 / 181 in ekos, but -57 / 90 in Indi Controll.
Then i entered 1 / 181 in indi control, and the scope slewed to -55 / 24 in ekos and 1 / 99 in Indi control.
At this point the scope was pointing North and to the ground. It did sound like the mount aborted movement in ra due to passing the meridian by more than 20%
Edit/ Can add that if RA was not stopped by the mount and could continue the rotation, then dec would be pointing at the correct altitude, with scope to east and counterweights to west...

Here is some images and log.

Slewed to 1 / 181 by selecting in kstars


Slewed to 1 / 181 by seting in Indi cp


File Attachment:

File Name: 2015-09-13-2-3.txt
File Size:53 KB
Proud owner of Observatory 17b - A remote Linux observatory.
Website: Observatory 17b
Build thread @ SGL: Starting summers observatory project
Last edit: 8 years 6 months ago by Magnus.
8 years 6 months ago #5055
Attachments:

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

Thank for your patience, it is higly appreciated!

Now this still baffles me, it just doesn't make sense. According to Celestron (www.celestron.com/c3/images/files/downlo..._nexstarcommprot.pdf)
So given that the initial values are Az = 0x4000 , Alt = 0x4000 which is btw is 1/4 of a full revolution ( 0x4000 / 0xFFFF = 0.25). So if you use their equation then the initial values are 90 and 90! So the azimuth value is being offset by 0x4000 to "zero" it.

In your last test, the initial azimuth value was 0x4000 and after slew is done, the value is 0x804D, which is almost double the value of 0x4000, so even though the azimuth move by 180 degrees, the celestron azimuth values only changed by 0.25 revolution, instead of the expected 0.5 revolution (since it moved from north to south). Directly converting (804D / 0xFFFF * 360) would yield an azimuth of ~180 but using the same equation on initial values result in 90 degrees azimuth. Of course altitude is another problem, but I might be missing something obvious here regarding how this should be calculated.
8 years 6 months ago #5062

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

Time to create page: 0.442 seconds