Polar Alignment | Astrometry style

[Gonzo]

thanks I'll figure it out.

[MikeP]

Hi Gonzo,

I'm not convinced by the method you are using but at the risk of making matters more complicated let me try and explain why.

The end result of polar alignment is for your mount's axis to be pointing at the NCP. If you have a mount with a built in polar scope, you need to be sure the polar scope is correctly aligned with the mount otherwise when you have finished, the polar scope will be aligned but the mount won't be.

The same is true if you put a camera on your mount and polar align by examining single images - if the camera is not aligned with your mount, the end result will be for your camera to be aligned to the NCP but not your mount. The misalignment may be small but it will be there.

There are software based methods for aligning in this way (PoleAlignMax being one). They get round the problem of the camera being misaligned by taking an image, plate solving, slewing a known amount, taking another image and plate solving again. If the MOUNT is misaligned the second plate solve will yield results inconsistent with the first plate solve plus the slew - the difference allowing the misalignment to be calculated and communicated to the user.

I'm not suggesting you try PoleAlignMax because it depends on your mount being controlled by computer - I'm just trying to use its method to explain why I believe the method you are using is flawed.

When you get your HEQ5, if you do hook it up to a computer, have a look at Alignmaster. It is easy to use and very quick to execute.

Good luck with the alignment experiments and your HEQ5 when you get it.

Mike

[Gonzo]

Thanks for everyone, I'll try that plate solving stuff whenever I got clear sky again

[Gonzo]

I have now started to write my own script to try Themos method.

#!/bin/sh

#shrink it a little for faster viewing...
convert $1 -resize 1024 npa.jpg

#generate wcs file,
solve-field  -B none -P none -M none -S none -R none -U none -T -p -O npa.jpg

#get Polaris & Lambda coordinates
plot=$(plot-constellations -w npa.wcs -B -L -v 2>&1 |  egrep "Polaris|Lambda" | grep -Ev The)

#filedate
FDATE=$(stat -c %n" "%y $1)

#plot-constellations
jpegtopnm npa.jpg > npa.pnm
plot-constellations -B -C -w npa.wcs -i npa.pnm -o npa_with_constellation.jpg  | grep -Ev "The|Part"

#paint artwork
convert npa_with_constellation.jpg -fill red -pointsize 16 -draw "text 5,20 '$FDATE'"  -draw "text 5,50 '$plot'"  output_$1.png

#show the result with imagemagick
display -title "Polar Alignment"  output_$1.png

What he does:

solve your picture using Astrometry and watermark it with the coordinates of Lambda UMi (λUMi) and Polaris (αUMi)

example:

This is just a start, planning to add a lot more things to the script at a later date.

In a nutshell, the script will ask for the two pictures you need to take according to Themos method, solve them, and generate a report

[themos]

Sweet! Are you using a Linux system or cygwin under Windows?

The tricky part would be deducing the position of your mount axis from the two pictures. This is the calculation done in cell C13 and C14 (also E13 and E14) of my spreadsheet. I am sure you can repeat it with a bc calculator script bc - GNU Project - Free Software Foundation . You would then annotate the image with both the position of the NCP and the actual place your mount is pointing, helping the user to visualise how they can improve the pointing.

Link to spreadsheet:

https://docs.google.com/spreadsheet/ccc?key=0Ags-dmFqKZThdHNxY3MxRzBlX0JJVFlURV8tbHFpVVE&authkey=CIyr34UJ

[Gonzo]

Sweet! Are you using a Linux system or cygwin under Windows?

I'm a Debian user, always been.

The tricky part would be deducing the position of your mount axis from the two pictures.

not that tricky, just some clever bash scripts

This is the calculation done in cell C13 and C14 (also E13 and E14) of my spreadsheet. I am sure you can repeat it with a bc calculator script bc - GNU Project - Free Software Foundation .

Link to spreadsheet:

https://docs.google.com/spreadsheet/ccc?key=0Ags-dmFqKZThdHNxY3MxRzBlX0JJVFlURV8tbHFpVVE&authkey=CIyr34UJ

I got your spreadsheet, will add the formulas into the script.

You would then annotate the image with both the position of the NCP and the actual place your mount is pointing, helping the user to visualise how they can improve the pointing.

That should be the end goal....

[themos]

I'd be interested in adapting what you come up with so that it can run under Windows. Thanks for sharing this.

[Gonzo]

Thanks for sharing this.

no problemo, I'll update this thread as I go along with my script.

I got some ideas.. now I just need to put them into a script

[Gonzo]

This is what I achieved using your method.

First, the numbers:

Then a 60" shot:

The new script is now bigger and better, can't post it as it's too big. I'll setup a quick web page on one of my domains.

Script output:

(it's jut the start....)

root@pig:/home/stephane/polaris/test2# ./polar.sh

Horizontal: horizontal.jpg

Vertical: vertical.jpg

Reading input file 1 of 1: "npa_horizontal.jpg"...

jpegtopnm: WRITING PPM FILE

Read file stdin: 1024 x 685 pixels x 1 color(s); maxval 255

Using 8-bit output

Extracting sources...

simplexy: found 5051 sources.

Solving...

Reading file "./npa_horizontal.axy"...

Field 1 did not solve (index index-219.fits, field objects 1-10).

Field 1 did not solve (index index-218.fits, field objects 1-10).

Field 1 did not solve (index index-217.fits, field objects 1-10).

log-odds ratio 178.493 (3.29922e+77), 26 match, 0 conflict, 14 distractors, 31 index.

RA,Dec = (18.2819,87.5435), pixel scale 68.5708 arcsec/pix.

Hit/miss: ++++++++++-++-+++--+-+-++++++-------++-+(best)------------------------------------------------------------

Field 1: solved with index index-216.fits.

Field: npa_horizontal.jpg

Field center: (RA,Dec) = (17.98, 87.55) deg.

Field center: (RA H:M:S, Dec D:M:S) = (01:11:55.618, +87:32:50.958).

Field size: 19.3075 x 12.9867 degrees

Creating new FITS file "./npa_horizontal.new"...

jpegtopnm: WRITING PPM FILE

Reading input file 1 of 1: "npa_vertical.jpg"...

jpegtopnm: WRITING PPM FILE

Read file stdin: 1024 x 685 pixels x 1 color(s); maxval 255

Using 8-bit output

Extracting sources...

simplexy: found 4328 sources.

Solving...

Reading file "./npa_vertical.axy"...

Field 1 did not solve (index index-219.fits, field objects 1-10).

Field 1 did not solve (index index-218.fits, field objects 1-10).

log-odds ratio 75.7954 (8.27018e+32), 13 match, 0 conflict, 12 distractors, 18 index.

RA,Dec = (103.54,87.7263), pixel scale 68.3553 arcsec/pix.

Hit/miss: -++++++-+-++++----+-----+(best)---------------------------------------------------------------------------

Field 1: solved with index index-217.fits.

Field: npa_vertical.jpg

Field center: (RA,Dec) = (103.2, 87.73) deg.

Field center: (RA H:M:S, Dec D:M:S) = (06:52:51.993, +87:43:48.215).

Field size: 19.1288 x 12.9037 degrees

Creating new FITS file "./npa_vertical.new"...

jpegtopnm: WRITING PPM FILE

root@pig:

output of the log files:

root@pig:/home/stephane/polaris/test2# more out-horizontal.log

Lambda UMi (λUMi) at (683.918, 310.483)

Polaris (αUMi) at (603.133, 288.023)

root@pig:/home/stephane/polaris/test2# more out-vertical.log

Lambda UMi (λUMi) at (665.011, 244.598)

Polaris (αUMi) at (636.24, 324.274)

root@pig:/home/stephane/polaris/test2#

[russellhq]

Great work Gonzo! Did this ever make it to Windows?

[themos]

It works fine under Windows/cygwin. I use it all the time with my EQ1.

http://code.google.com/p/eq-polar-alignment/