Cone Error? Or something else?

Hi all,

I was imaging the other night right next to Polaris (Declination 88 degrees).

I discovered after the Meridian flip that the image was no longer centred at all or even in the field of view.

Cone Error is the culprit I assume.
I have noticed slight frame rotation after meridian flips before, but in this case, once I got the target manually centred after the flip, the degree of rotation was extreme.

So much so that I had to crop an insane amount out of the stacked image.

I’ve attached a screenshot of the final stacked image so you can see how much rotation was involved.

The only thing is, I would have thought that the diffraction spikes would be misaligned with cone error and they are not.
I hope the axes on the mount are not the issue but I don’t know how to go about checking definively in any event.

Would appreciate any input on this as I’m sure there are many more experienced than me that might have encountered something similar.

Gear in question:

2600MC Pro

150PDS Newt

EQ6R-Pro

ASIAIR

📷 Screenshot 2026-04-15 at 14.47.06.jpg

I doubt that the mount could be off that much. When I have seen this it’s usually do to something in the imaging train getting loose and rotating.

When you say “Cone Error", I am assuming you mean cone error in the mount's orthogonality. Un-orthogonality in a mount will cause bad star alignment when the mount is flipped. You can test this by simply looking through an eyepiece visually after the mount is flipped. Plate Solving usually solves this problem and re-adjust the star alignment.

Mount Un-orthogonality should not cause rotation. For that to happen there needs to be rotation in the optical axis. More likely, something is loose in the optical train and the weight distribution caused by a meridian flip is causing the camera or something in the optical axis to rotate.

Lynn K.

Lynn K · Apr 15, 2026, 03:23 PM
When you say “Cone Error", I am assuming you mean cone error in the mount's orthogonality. Un-orthogonality in a mount will cause bad star alignment when the mount is flipped. You can test this by simply looking through an eyepiece visually after the mount is flipped. Plate Solving usually solves this problem and re-adjust the star alignment.
Mount Un-orthogonality should not cause rotation. For that to happen there needs to be rotation in the optical axis. More likely, something is loose in the optical train and the weight distribution caused by a meridian flip is causing the camera or something in the optical axis to rotate.
Lynn K.

Thanks Lynn & Tony.

I don’t believe there was any rotation in the image train as the Pre and Post meridian flip images were both perfectly flat-corrected using the same flats as I have been using for many months now(as I have left the camera rotation unchanged).

I thought that cone-error could cause rotation when very close to the NCP but could be mistaken

If this is something you are seeing everytime you filp and it’s not a one off you can always run this to measure and correct.

https://www.sharpcap.co.uk/conesharp

Tony Gondola · Apr 15, 2026, 07:04 PM
If this is something you are seeing everytime you filp and it’s not a one off you can always run this to measure and correct.
https://www.sharpcap.co.uk/conesharp

Thanks Tony. I see it frequently alright but the effect is very minimal. Imaging so close to the NCP I think exaggerated the effect. Will check that out thank!

Cone error can indeed result in field rotation when imaging close to the pole. This is because, near the pole, the dec. axis doesn’t move cleanly just in dec. As a result, pointing near the pole involves shifts in the dec. and ra axes to point at a specific target, and these shifts will result in field rotation. The mount can still point accurately if it has a good mount model and can compensate for this effect, but the field will need to rotate in order to point the same place after a flip.

Note that with cone error it is physically impossible to point directly at the pole. The cone of sky around the pole is completely unreachable, in terms of direct pointing.

This won’t be an issue with normal tracking as long as the mount is polar aligned, but a meridian flip would involve some field rotation.

The spider vanes should also show rotation (after alignment, because they are fixed in orientation relative to the sensor) and it’s odd that doesn’t show here in the stack, but my guess is there are fewer exposures after the flip and they were sigma rejected. A raw stack without rejection might show the rotated vanes, or just do a direct comparison of an exposure before and after meridian flip.

Frank

Freestar8n · Apr 16, 2026, 01:31 AM
Cone error can indeed result in field rotation when imaging close to the pole. This is because, near the pole, the dec. axis doesn’t move cleanly just in dec. As a result, pointing near the pole involves shifts in the dec. and ra axes to point at a specific target, and these shifts will result in field rotation. The mount can still point accurately if it has a good mount model and can compensate for this effect, but the field will need to rotate in order to point the same place after a flip.
Note that with cone error it is physically impossible to point directly at the pole. The cone of sky around the pole is completely unreachable, in terms of direct pointing.
This won’t be an issue with normal tracking as long as the mount is polar aligned, but a meridian flip would involve some field rotation.
The spider vanes should also show rotation (after alignment, because they are fixed in orientation relative to the sensor) and it’s odd that doesn’t show here in the stack, but my guess is there are fewer exposures after the flip and they were sigma rejected. A raw stack without rejection might show the rotated vanes, or just do a direct comparison of an exposure before and after meridian flip.
Frank

Thanks so much Frank. That clarifies a lot. I might try stacking without rejection and see what happens so!

I think polar alignment could also be a factor here, and you should make sure to align on the refracted pole, rather than the geometric one. It’s not a big effect unless you are at low latitude, but for best pointing and guiding you’d want it to be as accurate as possible when imaging near the celestial pole.

Normally the choice of refracted pole wouldn’t matter much, but it matters more here because you want to use the refracted pole so that rotations of the polar axis track a circle around the apparent pole in the sky. In SharpCap you have the option to “correct for atmospheric refraction,” and that is what you don’t want to do, because it will shift the polar alignment down to where it is geometrically, rather than where it actually appears in the sky due to refraction.

You can always check how much field rotation you have by aiming at a patch of sky around 88 degrees N and near the meridian, and take an image. Then do a flip and image the same patch of sky and see if it rotates. And you want to make sure you are centered on the same ra/dec because it’s the final corrections to get it centered that result in field rotation.

Of course, the simple solution for your imaging task is just to stay on one side and avoid a flip.

Frank

This is the problem with the ASIAIR, that it doesn’t re-center after the meridian flip…

Even when it was centred the frame was rotated due the non orthogonality of the OTA. I have looked into the geometry of it more since Frank explained the issue to me. The degree of severity is proportional to the reciprocal Cosine of the Declination angle so it is extremely pronounced at the pole and not so noticeable elsewhere

John Walsh · Apr 15, 2026, 01:51 PM
Hi all,
I was imaging the other night right next to Polaris (Declination 88 degrees).
I discovered after the Meridian flip that the image was no longer centred at all or even in the field of view.

Hello

How where these images guided ? OAG or guide scope ?

alpheratz06 · Apr 18, 2026, 03:45 PM
John Walsh · Apr 15, 2026, 01:51 PM
Hi all,
I was imaging the other night right next to Polaris (Declination 88 degrees).
I discovered after the Meridian flip that the image was no longer centred at all or even in the field of view.
Hello
How where these images guided ? OAG or guide scope ?

Guide scope. Guiding was fine all night. No problem on that front

John Walsh · Apr 18, 2026, 03:46 PM
819314819314alpheratz06 · Apr 18, 2026, 03:45 PM
John Walsh · Apr 15, 2026, 01:51 PM
Hi all,
I was imaging the other night right next to Polaris (Declination 88 degrees).
I discovered after the Meridian flip that the image was no longer centred at all or even in the field of view.
Hello
How where these images guided ? OAG or guide scope ?
Guide scope. Guiding was fine all night. No problem on that front

One possible issue in that case is a misalignment between guide scope and main tube : in that case, this may amplify field rotation due to a poor polar alignment. Rotation takes place around the guide star which generates bonheur translation and Rotation in the field of the main tube.

This error appears to happen simply by centering on the same RA/Dec with or without a meridian flip, so the guiding method isn’t playing a role. With a German Eq mount you can point to objects near the meridian in two ways, on either side of the pier, and imperfections in the mount model can induce some field rotation.

Cone isn’t really a mount imperfection and it can happen even with a perfect mount if the optic axis isn’t in the plane perpendicular to the dec. axis. The optic axis may not be centered in the tube, and the tube may not be perpendicular to the dec. axis. It means that rotations in the dec. axis will cause the optic axis to trace a wide cone in the sky instead of a great circle. Cone error is also called collimation error, and is another way the term, “collimation” is used in different contexts.

An issue with mount orthogonality would be an actual mechanical problem where the dec axis isn’t exactly perpendicular to the polar axis.

I think any of these issues could lead to rotation when pointing near the pole on either side of the pier. And any setup will have some amount of these errors, along with things like sag and polar misalignment, and they will all have more effect near the pole.

Frank

Related discussions