"Angle" definition for plate solving


Can someone help me understand how “Angle” is defined in SGP?

In Astrotortilla, you get a +/- angle and a yes or no “flipped”. How would that correspond to “Angle”, in SGP, which doesn’t take negative angles and doesn’t have a “flipped” checkbox?

E.g. AT says -128 flipped. What would I put (as an initial value until I get a successful solve) in SGP?

Also good to understand for aligning the Scope FOV displayed in Planitarium software like Stellarium.



There are 2 angles in SGPro. One is the positional angle which we don’t use much. It is simply the mechanical angle of the camera / rotator without regard to the sky. The one you are asking about… the one we sync to and the one reported on plate solves uses the sky angle reported north of east.


Well basically it sounds like that should never be entered by a human, just as the result of a plate solve.

It’s going to be dependent on where the scope is pointing.

I suggest you make It “disabled” for user entry. As it is now, it just adds confusion.


Not sure why you would suggest this… it is easily definable by a human since sky angle is standardized using degrees east of north. I probably should not have used PA in my original description as this is never exposed to the user… just trying to give you insight.


@Ken… Again you make us smile : :relaxed:

@dts350z says: As it is now, it just adds confusion. @Ken says: Not sure why you would suggest this

Yet in your posts, you say 1) plate solves uses the sky angle reported north of east and 2) since sky angle is standardized using degrees east of north

See…we hang on your every word here on this forum :smiling_imp:


If the mosaic tool sets the object to 0 degrees then the images can be taken a either 0 or 180 PA.
What do I set the flat on a future run since It could be rotated to either or these values?


Yes, my apologies… I did transpose those words.

This is a choice you make in the target settings. We do record the actual rotator PA of the images in the FITs headers. If you don’t want to have to worry about this, make sure ± 180 is selected for your targets so that the camera does not rotate after a flip.


I have the +/- checked but I am not sure what orientation this implies.

Example: I specify to rotate a object to 90.0 degrees manually or using the framing wizard.

Let’s say in my system , the images start on the east side with actual rotator position of 22.0 degree to get 90 degrees on the sky
It flips to west and the actual rotator stays at 22.0 degrees.

Now I want a flat and put in the object rotator dialog at 90 degrees ±180.

What position will it use?

22 or 202 on the actual rotator ?

Can I be sure that object at 90 degrees will match be the same every time?

I know I can check to the fits header to see the actually rotator position.
That does not help me to plan a exposure or a flat exposure.

I just don’t understand what ± 180 really means. Rather confusing


In my use case, with oag and a rotator, I just want the rotator angle set to exactly what I request - and ignore the image angle. On meridian flip that angle should change by 180 degrees exactly - and again ignore what the image angle is.

For me the oag angle is critical and accurately calibrated to the rotator angle - while the image angle is only approximate and could be different for different objects if I rotate the camera relative to the oag.

I’m happy to have the image angle stored in the fits header - but for flats and for finding the guidestar I need to set the rotator angle exactly, while the image angle plays no role.



All angles in SGP are Sky Angle, the exception to this is if you disable plate solving then you’ll get the rotator angle as we cannot compute a delta without a plate solve. The rotator angle is arbitrary and generally useless. Keeping the rotator angle at the sky angle means that it is an absolute value that doesn’t change. Which works out nicely as RA and Dec do the same thing. I still don’t understand why there is no “sync” for the rotator but that’s another story. So we internally keep an offset of the rotator sync inside of SGP and report sky angle.

First we’re storing an offset for you. So when you run a solve and since we store an internal offset for your rotator:
SkyAngle(90) = RotatorAngle(22) + 68.

So if you told SGP to rotate to 90 degrees it would go to:
90 - 68 = 22

Maybe? It depends on what you mean by “match every time”. This is only obscure because you’re using the ±180 option. If you want to guarantee that it always moves the rotator to 22 degrees then you should remove this option. Otherwise it can move between 22 and 202.

In short it means that for a merdian flip that the rotator doesn’t rotate and that SGP will go to the closest option of those two. Essentially it means that you are fine with either 22 (90) degrees or 202 (270) degrees, in your case. If you want the image orientation to always be the same you should NOT use the ± 180 option. For instance if you want to use the same guide star you should NOT use the ±180 option which will keep the Sky Angle the same. Forcing the angle to be the same will likely require some additional settings changes in PHD2 as the guide angle has now changed from what it was expecting after a meridian flip.

Hope that helps,


For me the sky angle is secondary and the rotator angle is primary - because the rotator places the guidestar on the guidere and is essential for autoguiding. The sky angle could be tilted or independently oriented for a given object for optimal framing - but that part is done manually for an object and is independent of the rotator/oag angle.

So if I start an image with the rotator angle at 22 degrees - that is what I would want to use as the reference - and when I flip I would set it to 202. And when I take a flat I would use 22 and 202. For me the 22 number comes from a planetarium field of view indicator and is part of the planning for an image.

It is also useful to have the image angle in the fits header and plate solved.

In other situations where the rotator is used for optimal image orientation - the image angle would be primary. But I think a large fraction of us use the rotator for guidestars - in which case we are not thinking in terms of the image angle - whatever it is - since it is some offset from the oag angle in general.



This thread didn’t help me much in understanding angle.
I’ve never bothered entering a value for angle because I don’t know how to find it or how to measure it.
Is there a visual tutorial that would help to explain what this is and how to find/measure it?



I think Jared answered my question.
That is ± 180 should be not be selected if I want to my flats to match for certain.

This option is for automatic 180 rotation usually for a OAG went it flips the meridian.


In what way do you think that Flats have to match Lights? Orientation of the camera or telescope should have no effect on how Flats are used with Lights. Flats are taken in the daytime, not at night. Associating Flats with flipping on the meridian is a disturbing concept to me. This may be something I’m not at all familiar with.




You might have some misconceptions about flats or you might be saying this stuff in a way that does not immediately make sense to me… Either way…

It does (can). Flats can remove dust motes that are close enough to show on the CCD (but not attached to the CCD). This means that rotation of the imaging train, in any way, after light frames are taken might result is less than optimal results. In addition to motes, the vignetting on each corner might no be equal with rotation… For you, it might now matter… for others it does. If this is something of concern to you ask SGPro not to rotate your camera after a flip… the images angle will change by 180 degrees, but stackers don’t care. If you need the camera right where it was (angle-wise) before the flip, choose “always”. This is not the default because it is really only needed for long FL OAG imagers.


Some refractors often have illumination well centered. So it does not matter that much.

Most compound systems and reflectors the center off illumination is almost always significantly off.

Test this if you think you flats work at any rotation.

Take a flat image an one angle then rotate 90 degrees iand take another image.
Divide one image into the other. The result should appear perfectly even across the image.



Thank you. I was completely wrong in my thinking - because I never rotate my camera I never experience that problem.



I have a square chip most of the time my orientation stays at zero degrees.