M1 testing the cc spacing

[Scott]

Managed a couple of hrs on the Crab last night with the 200p/baader cc and 414ex. While it's not terrible I'm still not sure things are right. could be tilt? could be spacing? could be flexure? I'm really not sure so any advice is most welcome

[Oddsocks]

Difficult one to analyse Scott.

I ran it through CCD inspector this afternoon but being a compressed jpg image there isn't an airy disk around the stars to produce a quality data set for CCD inspector to make an accurate measurement.

So the following is only a stab at it....CCD inspector shows a total tilt of 11% (in airy disk aspect, so this is not a huge easily visible thing, it is probably only a mm out of true), running from bottom left to top right, -145 deg from the horizontal. The arrow in the curve plot points the direction of tilt.

The 3D display shows a mostly flat surface but with both long edges turned down slightly which is rather strange and difficult to explain unless perhaps there is some slight guiding/tracking error along the same direction as the axis of tilt....Is the detector aligned horizontally to the RA axis or is the detector aligned with the direction of star elongation? If it was a spacing issue you would expect turned down (or turned up) edges on all four corners, not just the two edges closest to the optical centre.

I would look at tilt being the prime culprit, the stars are almost round in the centre and bottom left quadrant and progressively worse in the other three quadrants. The crab itself seems rather blurry for an Ha image so perhaps this is a mix of tilt and tracking error assuming the tracking is in the same direction as the indicated tilt but it could equally just be poor seeing conditions or focussing on a star in the bottom left quadrant or centre portion while leaving the rest of the star field defocussed due to the tilt.

I have attached the curve and 3D plot images, but wouldn't like to put too much faith in them, CCD inspector requires uncompressed images to work with, if you want to put up three or four FIT or TIF files in a Dropbox or Google Drive I, or another willing SGL member with CCD inspector, can run them again for you.

 

 

William.

[Scott]

William,
Thanks ever so much for taking the time to analyse this for me. My wife has control of my imaging laptop atm so I'll sort a few subs out tomorrow. I'm fairly sure the sensor is aligned with the ra axis but that too I can test tomorrow.

once more, Thanks for your time

[Scott]

@Oddsocks, here are a few fits from the run https://www.dropbox.com/sh/u32rj4np02tn3pu/AAArDaV_gW7eR4XHuHb8tZula?dl=0.

if you need anything else, let me know

[Oddsocks]

Hi Scott.

Below are the CCD inspector results. The plots are constructed from averaging all three fits simultaneously which improves the reliability.

I preset CCD inspector with your telescope focal length of 1000mm for the 200P and pixel size 6.45uM x6.45uM for the Atik 414EX giving a scale of 1.322 arcsec/pixel so the numbers that CCD inspector comes up with should be closer to real life.

As with the jpg result from yesterday, the curve plot and 3D plot show the sensor is tilted bottom left to top right relative to the optical axis, 6% at -162 deg from the horizontal but also both plots now show that the corrector spacing is out slightly.

Ignore the big dip in the middle of the plot, that is the presence of the crab in the centre of the frame, the important part is the overall shape of the curve and 3D plot showing some overcorrection of the coma corrector, unfortunately I don't have enough experience to say whether the spacing is too close or too far, unless another experienced CCD inspector user can say for sure you might have to try increasing and decreasing the spacing by around one millimetre, take some test shots and measure again.

The figures in the plots relate to percentages of FWHM, not physical dimensions so where the curve plot states the range of curvature in the image is 22.4% and tilt is 6% these are really very small quantities, you would be hard pressed to physically see 6% tilt in the system even though you can see the resulting tilt and curvature in the image.

CCD inspector can not tell you where the tilt originates, it might be mechanical in the focuser or the coupling between camera and coma corrector but one hint is the collimation cross-hairs in the curvature plot showing that the telescope collimation is out in the same direction as the tilt so I might be tempted first to go through the complete collimation procedure from checking the alignment and "squareness" of the focuser to the OTA, roundness and offset of the secondary and finally the primary adjustment, then carry out some more star tests to see what has changed.

To summarise, the system currently shows tilt at -162 degree to the horizontal, the CC distance is not quite right but only by a very small amount and the collimation is slightly off.

CCD inspector is quite a useful tool, it works well with rich star fields that don't contain large galaxy or nebula though it does ignore them to a certain degree.

If you want, post back links to future fits or tif files for analysis, three or four are enough (keeping the same angle on the OTA to prevent flex from skewing the results) after you have made some checks and adjustments and someone will run the files for you to see if things are improving...

William.