Jump to content


GoTo Mount error model

  • Please log in to reply
6 replies to this topic

#1 Misha


    Lift Off

  • -----
  • Posts: 17
  • Joined: 03 Aug 2008

Posted 23 August 2013 - 09:55 AM

The other night our local meet-up gathered to look at the Nova in Delphinus. My initial N-S position was probably quite a bit off, but I did a trick suggested to me by the person who sold me the scope: after I had the mount slew to Arcturus (first part of a two star alignment), I lifted the tripod legs enough to get Arcturus close to position in my red-dot finder. Then I proceeded as before. It worked well.

So then I started thinking about the alignment/calibration procedure. In theory, an EQ mount will track perfectly as long as the RA axis is perfectly parallel to the earths rotational axis. So what can go wrong?

1. RA Axis off
2. RA tracking motor inaccurate (including periodic errors)
3. Mechanical issues with the mount (like a slight twist in Dec or RA)
etc. etc.

Presumably, the software in NexStar+ use the alignment/calibration data to build an error model, and then uses this model to correct as well as it can for slewing to an object and tracking. Of course I realize that this model is not good enough for photography, hence the need for auto guiding.

What I wonder is if there is a comprehensice article somewhere that describes telescope mount tracking models. This would help me answer questions like

1. How much are things improved if I move the tripod legs after the first align slew?;
2. If my lat. setting on the mount is accurate, and the mount is N-S aligned, how close can I expect the first alignment slew to be?
3. How close do my Lat/Long settings need to be? Is it enough to select the city, or do I need to use GPS?

Can someone direct me to such an article if it exists?

#2 Cliff Hipsher

Cliff Hipsher


  • ***--
  • Posts: 1147
  • Joined: 31 Dec 2008
  • Loc: North Chesterfield, VA

Posted 23 August 2013 - 12:40 PM

AFAIK there is no such thing as a "tracking model". The purpose of a Polar Alignment is to align the RA axis to the North Celestial Pole. A Polar Alignment is purely mechanical and is done BEFORE any computer aided "alignment(s)". Once this is accomplished, tracking is simply a matter of moving the mount in RA at a rate of 15 degrees per hour. Any error in the Polar Alignment will show up as object drift. Because most mounts are not driven in DEC during normal tracking, any errors in the initial Polar Alignment will show up primarily in DEC. The more accurate the Polar Alignment is, the less drift error you will see. In an ideal situation your could do a Polar Alignment, use the hand control to slew to an object, center the object, and the mount would track the object all night long.

I can't answer #1. That is all trial and error. What I will say is that most GEMs have Azimuth and Altitude (aka Latitude) adjustment bolts that are used to move the mount head. IMNSHO, this is a much safer and more accurate method than lifting and moving the entire mount/tripod/optical tube as a unit..
The answer to #2 depends on how accurate your Polar Alignment (or as you call it the N-S alignment) is. Most latitude dials do not have enough resolution to be of any real use and are there just to get you in the ballpark.

For #3, the more accurate your time and position (Lat/Long) data is the better.

To sum up:

Polar Alignment is a mechanical alignment of the RA axis to the North Celestial Pole. This alignment affects TRACKING ACCURACY.

The computer "alignment(s)" (or more accurately "calibrations") is/are used to develop a POINTING MODEL, or a digital map of the sky. The computer in the hand control uses this digital map to point the optics at your selected target. Time, Date, and Location must be as accurate as possible. The pointing model has nothing to do with tracking. It does however, have everything to do with GoTo accuracy.

#3 rmollise



  • *****
  • Posts: 15404
  • Joined: 06 Jul 2007

Posted 23 August 2013 - 04:09 PM

What I wonder is if there is a comprehensice article somewhere that describes telescope mount tracking models.

Not sure what you mean by "tracking model." Whether your go-to alignment is good or bad, it is still at the sidereal rate. Error is determined by the periodic error in the mount. Accuracy by how close your polar alignment is. Go-to alignment has nothing to do with it.

#4 JAT Observatory

JAT Observatory

    NOT a Wimp

  • *****
  • Posts: 9428
  • Joined: 20 Feb 2005
  • Loc: In the Primordial Soup

Posted 23 August 2013 - 04:26 PM

AFAIK there is no such thing as a "tracking model".

While it's not a tracking model TheSky software has a module called Protrack. The downside is Protrack only works with Paramount scopes. It another way to get long unguided exposures.

Protrack uses the T-Point model to correct the scope's position during tracking. T-Point (and other point correction software) by itself only corrects pointing during the slew phase.

The problem is the pointing errors that are corrected by a pointing model during a slew such as flex can also occur when tracking. They just happen on a much slower time frame. But they are there and the amount of flex can change or increase during tracking but they are not corrected for.

So even if your scope has a perfect alignment it could still suffer from what appears to be tracking errors but are actually uncorrected flexing or other error build-up during a long tracking session.

I'm surprised a 3rd party pointing model software company hasn't created software like Protrack for use with non Paramount mounts.

#5 cn register 5

cn register 5

    Viking 1

  • -----
  • Posts: 760
  • Joined: 26 Dec 2012

Posted 24 August 2013 - 02:23 AM

If you want technical detail about mount models and errors two places to start are Toshimi Taki's web site, where he goes into considerable detail about mount models and errors, and Patrick Wallace's TPoint site. A search will find both easily. There are also university sites which have Astronomy course lecture notes on line.

The mount models in most mounts will handle the major errors, such as polar align error, cone error and the Hour Angle and Dec offsets. These are all that can be determined with a small number of positions.


#6 Misha


    Lift Off

  • -----
  • Posts: 17
  • Joined: 03 Aug 2008

Posted 25 August 2013 - 10:25 AM

Great! This is exactly what I was looking for. Thank you.


#7 cn register 5

cn register 5

    Viking 1

  • -----
  • Posts: 760
  • Joined: 26 Dec 2012

Posted 25 August 2013 - 05:13 PM

Thanks, it's good to hear that it helped.


Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics