Jump to content


Photo

Drift Error Calculator?

  • Please log in to reply
5 replies to this topic

#1 sternenhimmel

sternenhimmel

    Vostok 1

  • -----
  • Posts: 102
  • Joined: 05 Jun 2013

Posted 05 August 2013 - 01:17 AM

After replacing a few of the wooden pieces in my barn door tracker, I've been having tracking trouble, where stars are trailing in my photos. This is most frustrating as I had excellent tracking just a few nights ago, before I fiddled with anything (doh).

The rotation rate of the motors appears to be correct, and the swing arm distance hasn't changed. I use a laser for polar alignment, and the laser is aligned to the swing axis within 0.25 degrees if not less, so any polar alignment errors are small.

I was thinking of writing a program to analyze the images and, knowing (or calculating) the field position and orientation, the streak length and direction, and the exposure time, determine the source and magnitude of the error. Thus, if it's a polar alignment error, it would give information about how many degrees away from the celestial axis the alignment would have to be to produce the observe errors. Or, if the RA slewing rate is off, it could give precise values for this errors.

Does something like this already exist?

Here is an example photo I'd feed the program (1 minute at 105mm). Full version shows the tracking error:
http://i.imgur.com/4Al0OXk.jpg
Posted Image

#2 sternenhimmel

sternenhimmel

    Vostok 1

  • -----
  • Posts: 102
  • Joined: 05 Jun 2013

Posted 05 August 2013 - 04:32 PM

Well, I learned a few things about the trailing I'm seeing.

I compared frames captured sequentially, each a minute long, and realized there was some drift, but it was small (2-3 pixels) and nearly perpendicular to the trailing direction. However, I saw trailing in the same direction and length in frames I looked at.

Thus, excluding movement from shutter slap, the trailing I'm seeing must be some kind of periodic error that is a combined result of slight lateral play in the hinge I'm using (thoughts here?) and imperfect gear alignment.

Ugh.

#3 SMigol

SMigol

    Ranger 4

  • -----
  • Posts: 310
  • Joined: 30 Jul 2010
  • Loc: California, USA

Posted 05 August 2013 - 06:34 PM

You can stack your subs with Deep Sky Stacker with alignment turned off to see the trails -- or you can set the first image in the sequence as reference and then "compute offsets." The resulting values can be copied into a spreadsheet of your choice for analysis.

#4 sternenhimmel

sternenhimmel

    Vostok 1

  • -----
  • Posts: 102
  • Joined: 05 Jun 2013

Posted 05 August 2013 - 06:43 PM

Good idea! If the skies are clear tonight, I'll record some more comprehensive data to see if I can determine what's going on.

#5 SMigol

SMigol

    Ranger 4

  • -----
  • Posts: 310
  • Joined: 30 Jul 2010
  • Loc: California, USA

Posted 06 August 2013 - 02:46 AM

It works well even with uncalibrated files. Sometimes all I want to do is understand how drift or flexure is effected by what changes in hardware have been made. You can do this kind of testing under a full moon - just need enough stars to do some alignment.

#6 sternenhimmel

sternenhimmel

    Vostok 1

  • -----
  • Posts: 102
  • Joined: 05 Jun 2013

Posted 06 August 2013 - 04:07 PM

Thanks SMigol! I gathered a bunch more data the other night, and though I haven't exported the offset data, it does appear that tracking is pretty good, and that blurring of the stars is likely a result of vibrations, optical effects, and flexure. However, I will return here with offset data just to see exactly how poor/good my tracking was.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics