Goto Alignment

[gkec]

I have a CG5 mount woth Celestron Nexstar GOTO. In the light of recent discussions about alignment I made a note of what happened during the alignment process. I have an eg mount so levelled th mount and pointed it at due North. Once dark I set the mount to it's home position (pointing at the north polar axis) and centered on polaris ( I don't have a polar scope) using the alt-az adjusting screws.

I then ran through the 2 star alignment procedure using Capella and Aldebaran. Castor and a star I can't remember were used as calibrations stars. In all 4 cases I had to move the scope significantly to centre the object.

However once I had completed the process I then selected up Jupiter and it was bang on.

I strongly believe that the alignment process should be followed ignoring apparent errors. Black art processes such as releasing the clutches to centre the first star is just bonkers as that loses the North pole position.

[Stephen]

Not entirely sure what you're asking or saying here? If the alignment stars are significantly out, then your RA axis is pointing no where near Polaris. By using the handset and correcting each time, you are teaching the mount where Polaris (and by reference to alignment stars, the rest of the sky) really is and/or how far off you are.

Black art processes such as releasing the clutches to centre the first star is just bonkers as that loses the North pole position.

Umm, home position is effectively recalculated / reset every time you confirm an alignment star I think?

[cantharis]

Like you - I just follow the manual !!!

[gkec]

The way I see it is that there are two coordinate systems, the telescope's coordinate system and the "real" sky coordinates. The idea is then to find a transform function that will transform between the two coordinate systems. In a perfect setup the two coordinate systems will be identical and the transform function will be "unity". However this is hard to achieve. So the the transform function has to take into account three basic sources of error. (I am ignoring mount "levelling" errors in the east west direction.)

The telescope has a "home" position which is when the telescope position is 0^o RA 90^o Dec. When the scope is in this position it should be pointing towards the north celestial pole which is invariant and can be thought of as the sky home position. However getting this absolutely correct is difficult so there is a polar axis error which can be defined by a vector which defines a radial (RA) and lateral (Dec) component.

Then there is RA error which is a scalar and DEC error again a scalar.

So there are 4 values that have to be determined to be able to transform one coordinate system to the other.

Tracking to the first alignment star would give approximate RA/Dec error values some component of which would be polar alignment error.

Tracking to a second star fixes two know points in sky coordinates so you figure out the transform required to rotate the telescope coordinate system to match that of the sky so determining all 4 error values. So the scope's coordinate system has been aligned with the sky coordinates via the transform function. Further alignment/calibration stars just refine the transform function by making different measurements to remove measurement/observer alignment error.

So if you set the scope to home and align it as best you can to the north celestial pole. Track to the first star adjust with the handset to align the first star and the return to home the scope will return back to the same position it started from.

If you track to the first star undo the clutches to align it and then return to the home position the scope will return to a position that is not 0 RA 90 Dec and it will be pointing to a different bit of the sky. In effect there is a different polar axis error vector.

Both cases are a waste of time.

I also think there may be scaling factors to allow for small errors in the drive mechanisms.

[knobby]

said with tongue firmly in cheek ...there a lot to be said for a dob !

I have both and do find that the alignment process is very hit and miss, sometimes I get it right, sometimes not quite so right !

the dob ... just point it and browse

[chrisshillito]

If you track to the first star undo the clutches to align it and then return to the home position the scope will return to a position that is not 0 RA 90 Dec and it will be pointing to a different bit of the sky.

But will return to the correct position (dec=90, ra= any value depending on time) because the act of alignment has reset the coordinate system to ensure it. All that matters for the first alignment point is that the scope ends up pointing at it. It really doesn't matter how you arrive at that point. That on returning to 'home' it won't go back to your original starting position of no importance and simply reflects the fact that the original position was wrong .

The alignment system is not measuring your correction on the first alignment point. It is simply seeking to get one point fixed so that it can measure errors on the second and third stars.

Chris

[gkec]

But will return to the correct position (dec=90, ra= any value depending on time) because the act of alignment has reset the coordinate system to ensure it. All that matters for the first alignment point is that the scope ends up pointing at it. It really doesn't matter how you arrive at that point. That on returning to 'home' it won't go back to your original starting position of no importance and simply reflects the fact that the original position was wrong .

The alignment system is not measuring your correction on the first alignment point. It is simply seeking to get one point fixed so that it can measure errors on the second and third stars.

Chris

I see. So you ONLY release the RA clutch to reduce RA error. Which should make the locating of the second star etc more accurate thus making it easier for the observer. I can see this can possibly lead to better alignment as accumulated errors from moving the scope around to acquire and centre each star are reduced.

Have I got that right?