Q 1: Who is the target audience for this method?
A 1: Anyone who requires a good polar alignment in no more than 15 – 30 minutes.
By “good polar alignment” we mean one with a true NCP offset error of only a few pixels.
If you take long exposure captures, regardless of whether you work from a fixed or portable setup, a good polar alignment is critical to avoid field rotation problems.
Q 2: How long does alignment take with this method?
A 2: The very first time, it takes between 1.5 and 2 hours because the capture of the sky movement will need a long time to perform. However, once you capture this image, measure the starting and ending coordinates of several star trails and enter their data into the spreadsheet, you will have them for future use.
From then on, whenever you are in the same hemisphere and have the same optical train, accurate polar alignment will take no more than 15-30 minutes.
In this amount of time you should be able to do the following:
- Capture the “RA Mov.” image (3-5 min)
- Look and find in the “RA Mov.” image, the same two reference star trails as the ones you chose on the “Sky Mov.” image. Enter their starting and ending points coordinates into the spreadsheet and add the other 3 star trail coordinates (5-10 min).
- Use the correction shown in SV aligner spreadsheet to make the change to the mount (7-15 min).
In fact, if you understand how the SV aligner scale factor parameter works, there is no need to capture the long “Sky Mov.” image. Any starry sky rotation image around Polaris area will work. To learn about SV aligner scale factor, check the Annex 6 in the SV aligner quick guide.
Q 3: What exactly do you mean by same optical train?
A 3: Every optical component within your system where light is reflected or passes through. This goes from the first surface of your system the star light hits and ends with your camera CCD, both points included.
An optical train (A + a’ + a”) is equal to another (B + b’ + b”) only if all their elements are equal (A=B, a’=b’, a”=b”), are placed exactly in the same order and at the same distance (distance from A to a’ = B to b’; a’ to a” = b’ to b”).
Q 4: If you disassemble the optical train and mount it again with the CCD in a different rotation angle, do you still consider it as the same optical train?
A 4: Yes. No problem!
If you disassemble the optical train (telescope, filter wheel, focal reducers/flatteners, camera, etc.), store it and set it up again two weeks later in another place, it is considered the same optical train.
If you set up your system exactly as it was when you captured the “Sky Mov.” image, the focal length and optical corrections will not vary and the new “RA Mov.” image will be consistent with the former “Sky Mov.” image.
Different focus positions between the two captures are hardly perceptible compared with the total focal length of the system.
The CCD rotation angle has no impact on calculations. However, it may interfere in some way when you are trying to find the two reference stars chosen in “Sky Mov.” image within the rotated “RA Mov.” image.
In order to minimize the probability of a star chosen in “Sky Mov.” image disappearing from the image field in the “RA Mov.” image, we highly recommend selecting the two reference stars in the area within a square centered over the image.
(See step 3 at page 4. Quick Guide)
Once more, if you know how the scale factor parameter works, any “Sky Mov.” image where you can find the same reference star trails selected on your “RA Mov.” image will work. In this case the focal length difference between images will be shown in the scale factor value once you select the two reference star trails.
Q 5: Do I need to see Polaris in the CCD field?
A 5: No.
The only requirement is having curved star trails that enable SV aligner to calculate an accurate rotation center position. For example, if the mount is pointing at a position far away from the true NCP, you will not get star trails with the minimum recommended 60º curved paths.
This might affect the accuracy of the rotation center calculations.
(Note that 60º is not for just one trail, but for the total star trails shown in the image)
This does not mean that we could not try with images that show star trails covering less than 45º from side to side.
Remember that the shorter the total star trails become, the greater the potential for errors of calculation. The system’s focal length is closely related to the NCP offset at which we can point the mount while still having curved star trails.
As a general rule, you should point your mount to the area around the NCP.
Q 6: Is the accuracy of Park/Home position important?
A 6: No, for the same reason as the previous question.
We recommend capturing both images from Park /Home position to maximize the curvature of the star trails. If NCP is somewhere in the image, star trails will show a curve angle ranging from 90 to 360º.
When we draw back from the NCP, the star trail angle gets smaller in the image. At some point the star trails will look more like straight lines than curves. The Park /Home position therefore seems to be the optimal starting point in order to maximize star trail angles.
Of course, the degrees the star trails cover will depend on exposure time and selected speed in the case of “RA Mov.” image.
Q 7: Is it necessary to perform a polar alignment before using this method?
A 7: No, but it is highly recommended to perform a proper Polaris hour angle alignment with the polar scope.
If you do not, the “RA Mov.” image star trails may become very difficult to identify when compared with those in the “Sky Mov.” image. We tested and explained all of this in our post “Polar alignment (II): field trials”.
Q 8: Does the CCD axes have to be aligned with the mount axis?
A 8: It is not a must, but we recommend doing so.
If both axes are aligned with the mount axes it will be easier, more convenient and accurate when performing the mount corrections at the end of the process.
At that step of the process, when you know the required correction amount to apply to each axis, we recommend choosing a star near the NCP (a star within your CCD field while in Park /Home position) because the altitude and azimuth knobs will also produce movements aligned with the CCD axes. This makes the final adjustment step easier, more predictable and accurate.
Q 9: How many star trails segment have to be added to achieve an accurate alignment?
A 9: The first two stars trail segments of both images “Sky Mov.” and “RA Mov.” are mandatory and have to be the same stars on both captures, because they are the reference stars used for the alignment of the two images. Particularly for these two initial stars trails segments in each image, their starting points have to be measured accurately.
The extra star trails segment will help SV aligner to reduce the NCP offset calculation error.
That said, we have checked that with no more than 3 additional star trail segments (5 in total), the NCP offset estimation is highly accurate.
Note that, these extra star trails segments do not have to be the same on both images.
Q 10: Is it true that I can reuse the same “Sky Mov.” image for the same optical train as long as I am in the same hemisphere?
A 10: In short, yes it is.
The long and detailed answer is a qualified yes. It is true providing that you can identify the same two reference star trails in both images “Sky Mov.” and “RA Mov.” from your current (new) location.
If you cannot identify at least two star trails common to both images, then you will need to capture the “Sky Mov.” image again.
Q 11: Does cone error have an effect or need to be taken into account when using this method?
A 11: No.
The cone error, which results from the optical axis not being aligned to the RA axis, will affect the “go to” accuracy of the mount but has no effect at all on our method. You could even have a large image field offset between the two images and it would have no impact at all on our calculations.
The limits of the range of this image field offset will depend on the system’s focal length (approx. +/- 2º for a fl of 600 mm)
As a rule, if you can identify the two reference star trails on the two images, the calculations will work regardless of the cone error your system may have.