Help me test my Polar Alignment procedure

[themos]

Hello

I am developing a known but little used PA procedure for equatorial mounts without polarscopes but with a camera that can image more than about 2 degrees of sky (shorter side), an example being the EQ1 mount carrying a DSLR with a 55mm lens or a compact point-and-shoot. I assume that the camera will be connected to a laptop so that the user can quickly download and preview shots.

The idea is explained in ALIGN – A Polar Alignment Utility

but I will be implementing it in a Google Documents spreadsheet that we can all access from any platform.

Briefly stated, one takes just two images of the North Celestial Pole (NCP) area, records the pixel positions of Polaris and Lambda Ursa Minoris (the orange star on the "other" side of NCP from Polaris) in the two images, feeds these into the spreadsheet which responds with the desired pixel positions of Polaris. One then moves the mount so that Polaris is at those pixel positions (might take a few iterations and test shots to confirm).

To start off, you need to roughly align the RA axis and set your camera pointing to declination +90 or as close as you can get it (so that Polaris and Lambda will be in the picture). Lock the declination at this stage and don't change it until PA is finished. The two images will be taken at different RA angles, the first with the short side of the sensor horizontal and the second with the long side of the sensor horizontal. Review the two images and record the (x,y) pixel coordinates of the two stars. If anyone is happy to try that and report back with those 8 numbers, it will help exercise the spreadsheet.

Beta version of the spreadsheet is http://spreadsheets.google.com/ccc?key=0Ags-dmFqKZThdHNxY3MxRzBlX0JJVFlURV8tbHFpVVE&hl=en&authkey=CIyr34UJ

Not much doc with it, for now. The inputs are in light blue background and the output in green. The light yellow numbers are parameters that describe the relative position of the NCP to the Polaris-Lambda line and will need revising.

[themos]

I had a go with this procedure last night, between the clouds. I had just enough clear skies to shoot the two frames needed but not enough to implement the alignment solution.

I have now fixed the spreadsheet but I am still not sure if the relative position of the NCP (wrt Alpha and Lambda UMi) is accurate enough.

Anyway, here is a diagram of the two positions and a pdf snapshot of the spreadsheet.

Notice that all y-coordinates are given as negative numbers. This is because all image previewing software that I use have low y values at the top and high values at the bottom but the normal cartesian frame has y increasing towards the top. That means the image must be visualised as sitting entirely below the x-axis.

polar.pdf

[themos]

And here's a crop of one of the frames after it's been through Astrometry.net solver. The "Engagement Ring" is seen at top right (a rough circle of stars including Polaris) and orange star Lambda UMi is seen at bottom left.

[themos]

Not much interest, I see :-)

Are people happy with a less accurate PA when they don't have a polarscope or are there just very few with this type of setup? Or perhaps this method just doesn't appeal. I will be using it in the Med later in August so I'll know if it's any good.

[iamjulian]

Does sound interesting. Would be interested to hear about your progress, so keep posting

[themos]

I had a proper go last night. I took the two shots, ran the spreadsheet and iterated towards the indicated position of Polaris, taking shots and measuring each time. After about 6 steps, I was within 5 arcminutes of the Pole. I confirmed with a final couple of shots, measuring Lambda and Polaris and running the spreadsheet.

I then calibrated the RA rate with EQAlign and ExtraWebcam. EQAlign gives you a live drift graph so you can see the effect a twiddle of the pot makes.

I then took some subs of Lyra at 30 seconds, 60 seconds, 2 minutes, 3 minutes, 4 minutes and 5 minutes. I hit bad star trailing at 5 minutes but the 4 minute one was pretty good. Here it is, a 100% crop, there's M57 in there.

The only thing missing now is seeing if I can do the whole thing with the CLS clip-filter in place. I hope I will still get enough light through to use the software.

[themos]

BTW, I was so pleased I got this to work, I went and bought a license for ExtraWebcam.

[themos]

Here is a snapshot of the current spreadsheet.

You need to input all the fields in green background, for Mount Position #1 to begin with. The spreadsheet then calculates (in rows 7:11) where you should place Polaris or Lambda in either of the RA positions and reports your polar alignment error.

Once you do the mount adjustments, you can then measure and fill in Mount Position #2, to check on the new alignment. In this example, I have just put in the suggested values to verify that the calculations are stable.

Blue and orange are visual clues that the coordinates are for Polaris and Lambda (which is an orange star).

[themos]

If you have astrometry.net 0.25 sources on Linux, you can get that software to find and report the positions of Polaris, Lambda and NCP for you for each image you feed it.

First you need to add to the file brightstars-data.c the two lines

{ "NCP", "NCP",  0.0,  90.0,  2.0 },
{ "\xce\xbb""UMi", "Lambda UMi", 259.2367, 89.0378, 6.31 },

just under the Polaris line. Then run make to rebuild the software. It now thinks that Lambda and NCP are bright stars and so will report them when they are found in an image.

After doing solve-field NCP1.jpg (which produces the file NCP1.wcs), I do plot-constellations -w NCP1.wcs -B -L -v which reports

Trying to parse SIP/TAN header from NCP1.wcs...

Got SIP header.

Checking 5025 bright stars.

Polaris (αUMi) at (1319.39, 854.702)

NCP (NCP) at (1132.14, 1130.02)

Lambda UMi (λUMi) at (711.012, 1239.9)

Unfortunately, I can't run the astrometry.net software on the Windows XP laptop, so I won't be able to use that in the field. But it's a useful check on the spreadsheet calculations.

[LaurieY]

This is an interesting idea. I can't easily get 2 degrees of sky in my setup, I don't want to take my camera off the scope onto a lens but I suppose the idea could be extended to encompass 2 x 2 images with astrometry providing the values for the gaps between the images. I have read somewhere that astrometry.net does build & run under cygwin, though I haven't tried that. I run V0.34 under a Linux VM on my windows box with no problem.

Laurie