11 replies to this topic

[pgseye]

I suspect I am overthinking this. I recently bought a SA 2i and note the conventional way to PA is with the inbuilt polar scope. Obviously this ensures the mount is properly aligned. However, I also bought an ASIAir and guidescope/cam which I plan to use for PA instead. It occurs to me that PA using a scope really requires that the scope is perfectly aligned with the mount - i.e. if the scope axis is not parallel with the mount axis, attempting PA is pointless, isn't it? So, do people just eyeball the alignment of the scope and mount, and that is good enough? Or are there some other steps to follow to ensure the scope axis is as parallel to the mount axis as possible?

 

Perhaps overthinking it as I said...

 

Thanks,

 

Paul

[Sacred Heart]

Paul,

 

Are you using the mount like this??

 

https://www.youtube....h?v=oPluErQoOpY

 

If you are using it as an equatorial, like the video with the DSLR, mount the guide scope like the DSLR in the video. Do your PA then remove the guide scope replace with whatever you are using.  If you intend to guide, the guide scope does not have to be straight down the center, it can be mounted on the side of a DSLR. as long as it is pointed in the same direction.

 

I hope this helps,   Joe

[pgseye]

Thanks Joe,

 

Yep, I've mounted the guide scope on a vixen rail attached to a lens collar like so:

 

 

Attached Thumbnails

• 

[prookyon]

This kind of polar alignment is not just trying to center the celestial pole in the photos (like you would do through the polar scope) - the process is completely different.

It takes several photos of the sky with the RA axis rotated between the photos. The photos are platesolved which gives the exact coordinates the camera was pointing to. Based on those coordinates it's possible to determine where exactly the RA rotational axis is pointing to (since the only movement was on that axis). If the mount was perfectly polar aligned it would point to celestial pole.

That rotation makes it irrelevant if your camera is not parallel with the axis.

Don't use AsiAir myself, I know some software requires to at least point in the general direction of the pole, some work in any area of the sky.

[bignerdguy]

I use a program called SharpCap pro to do software PA.  The ASIAir works similarly to that program.  it doesn't require the scope to be 100% perfect as it uses platesolving to figure out where it is pointed and then uses rotation to determine how far off the mount is pointed from the Pole.

[pgseye]

This kind of polar alignment is not just trying to center the celestial pole in the photos (like you would do through the polar scope) - the process is completely different.

It takes several photos of the sky with the RA axis rotated between the photos. The photos are platesolved which gives the exact coordinates the camera was pointing to. Based on those coordinates it's possible to determine where exactly the RA rotational axis is pointing to (since the only movement was on that axis). If the mount was perfectly polar aligned it would point to celestial pole.

That rotation makes it irrelevant if your camera is not parallel with the axis.

Don't use AsiAir myself, I know some software requires to at least point in the general direction of the pole, some work in any area of the sky.

Thanks.

Perhaps I don't understand things well enough. My thought in all of this is that in the initial instance of polar aligning with a camera (that is essentially freely attached to a mount), rather than a polar scope accessory (that assumes a fixed position within the mount), you are susceptible to propagate alignment errors in your workflow if your scope and mount axes aren't completely parallel.

I'll need to think about this...

[prookyon]

Maybe a simple way to think about this would be following:

Let's say you have something on an equatorial mount - camera, telescope - doesn't really matter.

It can be pointed to anywhere in the sky.

You then somehow determine exactly where in the sky it is pointing. You get some RA and Dec coordinates.

Now you rotate the mount RA axis (and nothing else!).

As a result you are pointed to somewhere else in the sky.

If your mount was in perfect polar alignment then the result of this rotation was that *only* the RA coordinates are different now. It should be quite obvious since you only rotated the RA axis.

But what if you determine that the Dec coordinate also changed? Well I guess the mount was not perfectly polar aligned then.

In simplified terms this is what is happening in such platesolving based PA.

[BucketDave]

Effectively, there are two sorts of error at work.

Polar alignment error - the mount's RA axis is not aligned with the Earth's axis.
Cone error - the telescope is pointing a few degrees away from where the mount is pointing.

The two are independent, so you can do PA by plate-solving just fine, even if you have a bit of residual cone error (which all of us do, more or less).

[matt_astro_tx]

Thanks.

Perhaps I don't understand things well enough. My thought in all of this is that in the initial instance of polar aligning with a camera (that is essentially freely attached to a mount), rather than a polar scope accessory (that assumes a fixed position within the mount), you are susceptible to propagate alignment errors in your workflow if your scope and mount axes aren't completely parallel.

I'll need to think about this...

Prookyon is doing an admirable job explaining it, but I'll throw in my 2 cents and maybe it will help to have a second perspective.

 

When you're polar aligning using a polar scope, you're doing what's called "bore sighting" the mount.  You're literally looking down the barrel of the Right Ascension axis and eyeballing its alignment with the north celestial pole.  For this process to work the polar scope must be in precise alignment with the rotational axis of the mount so that when complete your right ascension rotates parallel to the earth's equator.  Any misalignment between the polar scope and the RA axis will result in a polar alignment error because your pointing will be off center.  Oh and you must be able to see the celestial pole from your location in order to use the polar scope!

 

When you use a software program like ASIAir and a camera to do this however, you aren't bore sighting (looking down the barrel).  Instead, the software will have you take a picture, then rotate the RA axis a bit and take another picture, and sometimes do it again and take a third picture.  This typically works best with your telescope pointed East or West, NOT toward the pole as you might expect.  The computer then plots the arc along which these pictures were taken on the celestial coordinate grid.  Knowing that this arc represents the rotation of your right ascension axis it can then do some complex math (mainly spherical trigonometry) to precisely determine where the center of the rotational axis is pointed.  By comparing where you are pointed vs. where the celestial pole is located, the computer displays your pointing error and can tell you how much and in which direction to move the base of the mount to achieve precise alignment.  You then adjust the mount slightly and take another picture, and the computer updates your pointing error.  You repeat this process until the error is effectively zero.

 

I also use the ASIAir and an SA 2i and polar aligning in this manner is a thing of beauty.

Edited by matt_astro_tx, 25 April 2025 - 04:20 AM.

[pgseye]

Effectively, there are two sorts of error at work.

Polar alignment error - the mount's RA axis is not aligned with the Earth's axis.
Cone error - the telescope is pointing a few degrees away from where the mount is pointing.

The two are independent, so you can do PA by plate-solving just fine, even if you have a bit of residual cone error (which all of us do, more or less).

Thanks,

 

I guess cone error was what I was meaning but didn't know the term for.

 

So the error in my thinking is my erroneous assumption that the scope and mount axes must be parallel. I guess that now makes sense. So am I getting closer to a light-bulb moment in the following thought process:

 

If you have perfect polar alignment (and no cone error) taking an image, plate-solving, rotating however many degrees along RA, taking another image and plate-solving, would have both images pointing exactly at the celestial pole.

 

But if you have imperfect polar alignment (+/- some cone error), the above will give different plate-solved coordinates, because you're pointing to two different bits of the sky.

 

However, this doesn't matter, because the software can use this information to tell you how to adjust the mount to obtain alignment even if your scope isn't pointing directly at the celestial pole.

 

Is that about right?

 

(My brain is probably regurgitating some of what you good people have already written above, but I'm just working through the thought process...)

[wongataa]

If you have perfect polar alignment (and no cone error) taking an image, plate-solving, rotating however many degrees along RA, taking another image and plate-solving, would have both images pointing exactly at the celestial pole.

 

Yes

 

 

But if you have imperfect polar alignment (+/- some cone error), the above will give different plate-solved coordinates, because you're pointing to two different bits of the sky.

Yes

 

However, this doesn't matter, because the software can use this information to tell you how to adjust the mount to obtain alignment even if your scope isn't pointing directly at the celestial pole.

 

 

Correct

[matt_astro_tx]

If you have perfect polar alignment (and no cone error) taking an image, plate-solving, rotating however many degrees along RA, taking another image and plate-solving, would have both images pointing exactly at the celestial pole.

Not quite.  Neither image will "point exactly at the celestial pole", and that's not what we're concerned about.  We are concerned with the rotational center of the mount's right ascension axis pointing exactly at the celestial pole.  If we achieve this and have perfect polar alignment (and no cone error) then, all things being equal, we can set the mount to track at sidereal rate, point the telescope at a target, and it will not drift but will stay perfectly centered in the field of view for the entire night.  This is what allows us to take long exposures without stars trailing.

 

But if you have imperfect polar alignment (+/- some cone error), the above will give different plate-solved coordinates, because you're pointing to two different bits of the sky.

Plate solving is the process of matching an image to the stars in the sky.  So, no two images will have the same plate solved coordinates, unless they're taken in exactly the same spot!  Rather, if you have imperfect polar alignment, when you place a target in your telescope's field of view it will slowly drift out of view with time.  And if you were to take a long exposure image, the stars would have trails behind them.  This is what we want to avoid.

 

However, this doesn't matter, because the software can use this information to tell you how to adjust the mount to obtain alignment even if your scope isn't pointing directly at the celestial pole.

This is correct.  The software uses the plate solved coordinates of the images you take during polar alignment to determine the mount's offset from the celestial pole, and in turn tells you how far to move the mount.

 

Hope that helps!