×

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

Bi-monthly release with minor bug fixes and improvements

Driver OnStep (LX200 like) for INDI

  • Posts: 322
  • Thank you received: 31

That is still a planned feature, not an actual one. The actual one in the April 1st release is a Light Pollution Eliminator!



At one point, I was reviewing the spreadsheet , and noticed that 'Steps per second' is calculated as:

(Steps Per degree / (60 * 60)) *15

I thought that it should be 15.041066, because:

Sidereal day length = 23 hrs *60*60 + 56 mins *60 +4.1 seconds = 86,164.1 seconds

And 15 seconds would be 15.041066

When I asked Howard why he is using 15 instead of 15.041066, he said that the calculations use sidereal seconds already, not solar days, hours, minutes, or seconds.

This may be the source of the discrepancy that James is seeing?
The following user(s) said Thank You: Alain Zwingelstein
Last edit: 5 years 11 months ago by Khalid.
5 years 11 months ago #24819

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

  • Posts: 161
  • Thank you received: 39
kbahey: Basically nope. The part that had the really minor error (THe 1.002... number) was actually what corrects for it internally. It's not the error I'm seeing though. (The error I'm seeing while small is a lot bigger than that adjustment!) I did have some great 180-300 second exposures (It was good, by the time I'd moved my computer out there... it was partly cloudy, so not great exposures of anything but general stars.) Did have a 100% error with RA (Shaft coupler came undone.) I thought I'd fixed that a while back, but I'm fairly certain that's not what was causing the issues.

Still can't find any reason why it should have that, at all. As far as I can tell from the INDI driver and OnStep code, there doesn't appear to be anything. Explaining it.

Definitely saw the focuser want to take a kzillion photos. As far as I can tell from a quick glance, that's an EKOS/Kstars problem. I'm guessing a loop in there somewhere, checking for a state, and firing off a new one as fast as it can. (Sometimes it was 1, sometimes 6+ attempts to get an image.)



Sheesh, you've already got an LPE going? They've always had manual ones, but I do like the idea of automating it. I'm behind on hooking up a PEC sensor to mine. I still haven't build the LPE, I think I'll use a relay instead of the plans as suggested. As it is, it seems, excessively safe.
5 years 11 months ago #24823

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

  • Posts: 452
  • Thank you received: 71
James,

If the tracking is not good most probably the Autofiocus will fail.
But As far as I tested, auofocus odes fail after some pictures, not running and endless loop.
May be Blueshawh has more experience with Autofocus (I just tested it because he reported on that and for my own curiosity).

Do you autoguide with ST4? May be there is something on this side! I eperienced erratic corrections because I forgot the Pullup in config ....

I do not understand the "Shaft coupler came undone" (Google translator ...)
5 years 11 months ago #24836

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

  • Posts: 322
  • Thank you received: 31
James,

Did you check the basics? Specifically : Polar Alignment, and Balancing, specially in RA?

I've had lots of problems due to these two recently (on a non-OnStep mount).
5 years 11 months ago #24838

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

  • Posts: 257
  • Thank you received: 22
Autoocus needs to be able to perceive(high enough snr) a star to focus. If it's way out it can't figure out which way to move. I think high backlash, as in the case of my o-ring/knob drive SCT version, can also cause fails in the algorithm. Very unstable skies can also cause it to get confused. This seems like fairly normal robot activity to me. Handy, easily confused, sometimes comically so.
Last edit: 5 years 11 months ago by Ray Wells. Reason: Friday
5 years 11 months ago #24847

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

  • Posts: 161
  • Thank you received: 39
The tracking/alignment issues isn't related to the focuser. Even with it being seemingly off, it only shows up visibly on 60+ second photos, the up to 5 second exposures (usually less, but one of my testing sessions was through wispy clouds) for tracking show no elongation, and unless I'm really out of focus, it can easily lock onto a star.

I agree about the possible backlash. Sometimes, when testing, it decided, when almost in great focus to go off to never find a star land, which I'm pretty sure is backlash. (I may switch from the gear setup to a belt setup. Or use 3d printed gears, instead of laser cut ones, allowing less possible play on the conical shaft. Right now I know it has a little bit of play.)

This is the focusing issue, It works, and yes sometimes is a little confused. The issue is the multiple exposures being requested:

Here's an example, from a log:
[2018-04-06T03:17:03.727 CDT DEBG ][ org.kde.kstars.indi] - LX200 OnStep : "[SCOPE] RES <0> "
[2018-04-06T03:17:03.727 CDT INFO ][ org.kde.kstars.ekos.focus] - "Capturing image..."
[2018-04-06T03:17:03.728 CDT DEBG ][ org.kde.kstars.indi] - LX200 OnStep : "[SCOPE] CMD <:GR#> "
[2018-04-06T03:17:03.728 CDT INFO ][ org.kde.kstars.ekos.focus] - "Capturing image..."
[2018-04-06T03:17:03.729 CDT INFO ][ org.kde.kstars.ekos.focus] - "Capturing image..."
[2018-04-06T03:17:03.729 CDT INFO ][ org.kde.kstars.ekos.focus] - "Capturing image..."
[2018-04-06T03:17:03.732 CDT INFO ][ org.kde.kstars.ekos.focus] - "Capturing image..."
[2018-04-06T03:17:03.733 CDT INFO ][ org.kde.kstars.ekos.focus] - "Capturing image..."

It will continue, but only after all of those have completed. (Usually it only takes one, but sometimes... there's been a whole screen in small font in my text editor of them.)

I've filed a bug in KDE for kstars, as I think that's where the problem is. Often it works pretty well.
---
As far as the tracking issues, I probably need to double check those, but usually it's seemed pretty good.
5 years 11 months ago #24964

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

  • Posts: 257
  • Thank you received: 22
Just getting going tonight, finally some stars! As I was getting focused I noticed mine was double exposing. After a while I noticed the "frames" field was on 2. this is the number of frames to average(live stacking) before deciding focus values. If this is a high value it might explain the weird multi exposures.
Hope this helps!
5 years 11 months ago #24971

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

  • Posts: 161
  • Thank you received: 39
It was a problem, and it's fixed in most versions marked as 2.9.4 as far as actually triggering multiple times. So that's great. There's the question of why it's called multiple times, but with the code change, it doesn't matter.
5 years 11 months ago #24987

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

  • Posts: 452
  • Thank you received: 71
Hi James,
Your implementation for the focuser works fine as far I could verify and also Autofocus works great.
The anoing thing is that when I have a second focuser in the firmware then I have two different implementations and that's really confusing.
I would prefer to go a step back, removing all code about the second focuser so that we have a clean code again before asking for merging in Master.
Are you OK with that approach?
5 years 11 months ago #25401

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

  • Posts: 161
  • Thank you received: 39
I've been working on PI Camera stuff for a bit so, I haven't gotten a change together yet.

I looked a little and I think it can be cleaned out, I've got a TODO around renaming OSFocus1RateSP? I'd tacked on the re-initialize the focuser position (for min/mid to it) The saving position isn't great for the way I use it. Put the camera in (repeatable), then have it move out.

Other than that, I think we can remove the rest of OSFocus1 stuff without issue.

Might not be bad since I haven't looked at the secondary device for Focuser2 to just throw it on a FOCUS2_TAB

There is the outstanding issue of it calling capture multiple times (not much of a problem on Kstars > 2.9.4 but I need to look into why. See KDE bugreport earlier.)
5 years 11 months ago #25489

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

  • Posts: 452
  • Thank you received: 71
James,

I will try to get things cleaned up, just want avois we both work on the same thing :-)
I also thought about leaving Focuser 2 in a FocusTab2 until we find a method of binding two focusers in via the focuser interface.
For KStars I never noticed ... I a using dayly build so now am on 2.9.5 already.

Happy with the PI camera? Do you intend to use it for guiding?
5 years 11 months ago #25549

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

  • Posts: 161
  • Thank you received: 39
I'm kinda meh about it. I'd love to have it working for guiding, I've gotten it to work once for guiding. (On Sirius?, I can't recall) However, it doesn't work well, at least with lin_guider (thus the attempt to get it to work natively with INDI and it's guider.) I got a NexImage 5 (Quite cheaply), which seems working ish. I discovered it would lock up, with other stuff last night, so I'm not sure exactly why. I had run tests of it just being on for probably hours watching some flowers in my yard. Frustrating is a word for it. My guess is that it's not sending the frames or something. I have logs to look at. Wheeeeee!

One thing that was frustrating was with the current OnStep Alpha and current indi seem to have slow serial connections. Like seconds long delays. I can go into a serial terminal and type some commands and it responds (Eg: :Gt# (time) type) pretty close so I'm sure it's not the serial connection itself. This is on a Mega with a DS3231 Focuser and RA/DEC, ST4 was set to pullup. A testing Mega (nothing on it) didn't have that delay problem until I flashed it. So not sure what's up. If anyone wants to take a look I've got 37MB of logs... and no recollection of when I had the issues. Everything seems to respond near instantly in the logs, that I've seen so far. when going through them, does anyone else get that?
5 years 10 months ago #25825

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

Time to create page: 0.655 seconds