CGEM: drift alignment vs 2 star alignment

[Florin Andrei]

Let me see if I get this right. With regards to polar alignment, the correct procedure for the CGEM is:

1. Do a rough alignment with the polar scope.

2. Turn the mount on, select Quick Alignment, which basically does nothing.

3. Calibrate, then do drift alignment in PHD2. This will aim the RA axis straight at the North Pole with good precision.

4. Then do the CGEM 2-star alignment (or 2+4 star alignment), so that goto on any object becomes very precise.

Is that correct?

Or does step #4 affect the precision obtained during the drift alignment in any way? I know step #3 involves physically moving the RA axis closer to the pole, but I wonder if the CGEM actually applies "corrections" in software to that position during step #4?

What's the best procedure so that polar alignment is as precise as possible, while goto precision is also good enough?

[Ken Self]

After Step 3 I usually do another calibration if the adjustments were significant. 

Apart from that step 4 will not affect tracking.

Personally I don't bother with step 4. I just plate solve.

[mahaffm]

I might be wrong but does there need to be a Step 2.5. . . Slew the mount to Celestrial Equator and near the Meridian.

I didn't think PHD2 can perform calibration unless the mount/scope is pointing to the south near the Celestrial Equator and near the Meridian, for folks in the Northern Hemisphere.

Mark

[Andy Galasso]

On Tue, Sep 26, 2017 at 12:13 PM, mahaffm <bigiri...@gmail.com> wrote:

I might be wrong but does there need to be a Step 2.5. . . Slew the mount to Celestrial Equator and near the Meridian.

I didn't think PHD2 can perform calibration unless the mount/scope is pointing to the south near the Celestrial Equator and near the Meridian, for folks in the Northern Hemisphere.

Right, step 3 "Calibrate, then do drift alignment in PHD2" would imply slewing to near the meridian-celestial equator intersection (drift align instructions: azimuth drift position)

(Technically, phd2 can calibrate at any location other than right near the pole, but for best results the recommendation is to calibrate near the meridian-celestial equator intersection.)

Andy

[Florin Andrei]

Right, I calibrate near the 0 / 0 coordinates, as shown in the documentation.

So, is that the consensus? Should I just use a plate solver such as PlateSolve 2, etc? I wouldn't mind that, I'm just trying to figure out a procedure that provides the best results overall.

Am I correct to assume that doing the 2+4 star alignment in CGEM after I've done drift alignment in PHD2 may apply software corrections that may diminish the precision of the polar alignment done in PHD2? It's not entirely clear to me how the 2-star (or 2+4 star) alignment works, but if it tries to determine the actual orientation of the RA axis and the Dec offset (and who knows what else) from the images seen in the reticle, then it might as well think it knows better and do the math for a slightly different polar axis, depending on the results of the alignment.

It seems to me that for best results, polar alignment should be done via the drift alignment method (or some other high precision technique) - and then the mount should be told "hey, these are the RA and Dec origins / offsets, don't bother trying to deduce the RA axis orientation, because that's done already". Does that sound right?

[mahaffm]

The NexStar HC star alignment, (2 star, 2+4, or other) does not change the mount polar alignment. The NexSar alignment simply aligns the mount to its surroundings by creating a sky model based on your selected stars for pointing accuracy. When you tell the mount to go to a location in the sky the mount uses the sky model that was stored in its memory from the star alignment process. I believe the alignment process also corrects for Cone error. As long as you DO NOT mess with the ALT or DEC knobs and simply use the NexStar HC for your star alignment your polar alignment will not change.

Hope this helps,

Mark

[Florin Andrei]

Perhaps it's my fault, I am not getting the point across properly. Yes, I realize the star alignment done on the hand controller does not physically alter the polar alignment, since there are no servos on the RA and Dec knobs. I am just trying to figure out how does the controller deal with errors in polar alignment, whether it can actually detect them during the star alignment, and whether it tries to compensate.

Let's say you do polar alignment with the polar scope. It's close, but no cigar.

Then you do the full 2+4 star alignment with the hand controller. At this point, the HC has enough information to estimate not only the various offsets, but also the real orientation of the whole mount, relative to the celestial sphere. It could probably estimate PA errors too, I presume. Does it apply corrections based on that estimate?

Or does that become part of the cone error correction? In other words, does the HC think "PA is assumed perfect, all errors are due to cone error", or does it think "I am detecting PA errors, let me estimate the real orientation of the RA axis and use that instead"?

[Kevin S]

The mount does estimate polar misalignment after the 2+4 star calibration.   (align -> polar align -> display align) will show you the offset, but does nothing with it regarding tracking or goto slewing.   PA matters for field rotation...for the most part, nothing else.   The RA axis is calibrated by the 2+4 calitration....not PA.   And...there are no RA and DEC knobs....they are Altitude and Azimuth.  they are not the same thing.

Regards,

Kevin

[mahaffm]

Hey Kevin,

Thanks for clarification on the Altitude and Azimuth, I'm not sure what I was thinking when I typed that! With our recent string of clear nights, imaging late into the morning hours and going to work without much sleep I think is catching up to me.

Kind Regards,
Mark