still struggling with this

[brrttpaul]

Took lights darks bias and flars everything seemed ok , opened all the files and calibrated then stacked then went through startools and the lightroom just dont understand why im getting the blue and red squiggles, any pointers? im not worried about the colour balance at the moment thatll come

[steppenwolf]

These 'squiggles' are particularly hot pixels and they form a line because you have some image drift. Darks are a good calibration tool but will not remove all hot pixels from your images. When you stack your images you need to use a stacking method like SD Mask or Kappa Sigma as these methods remove 'outlier' pixels.

[brrttpaul]

I used the sigma clipping, I never seem to get it using DSS only maxim. I have cleared all my flats darks etc and will have another go tonight

[VilleM]

Problem here is that your camera is not cooled (Im assuming youre using un-modified D7200). Temperature variation ruins dark frames, I wouldnt use them at all with un-cooled DSLR. Instead stack with methods mentioned above and use dithering.

-V

[brrttpaul]

im using a starlight xpress sxvhf9 osc

[steppenwolf]

I used an SXVF-H9C with MaxIm DL for many years and found that the combination worked very well indeed. With the amount of drift that you have, you have in effect got 'dithering' between your images so using Maxim DL, SD Mask should work very well. Try the following settings which work very well for me. Just to be clear, you are calibrating BEFORE you stack?

SD Mask Settings In MaxIm

 

[brrttpaul]

yes im calibrating before, My scope is permanently set up so going to try and solve one thing at a time (probably do the drift alignment to see how far out i am first) depending on the time i will try and take som subs then,  looking at that the only difference is i dont use IEEE Float and number of passes is 5 i think but will check now. EDIT just checked my sigma factor was 3.00 my normalization was none and ignore black pixelss was ticked

[alacant]

20 hours ago, steppenwolf said:

you have in effect got 'dithering' between your images

Hi. JTOL and please correct me if I have misunderstood. The hot pixels are creating the lines because the op has constant drift. To dither, the sensor position must be moved at the end of the exposure only, not constantly during it. HTH.

[brrttpaul]

just wondering could it be a camera problem? do starlight xpress service cameras?

[ollypenrice]

On 29/08/2016 at 13:12, steppenwolf said:

These 'squiggles' are particularly hot pixels and they form a line because you have some image drift. Darks are a good calibration tool but will not remove all hot pixels from your images. When you stack your images you need to use a stacking method like SD Mask or Kappa Sigma as these methods remove 'outlier' pixels.

Point taken, but that's a lot of drift, surely? My question to the OP would be, 'If you stack the first and last frames, how far apart are the borders of the image? If they are as far apart as the length of the squiggles then, yes, the drift has created them. They just look a bit long to me.

Olly

[brrttpaul]

I dont think its drift TBH as i get them on single subs also (not always but more often than not), im just trying drift alignment see how far out I am, polar alignment error is 1.6 arcmin,  RA rms is 0.04, DEC rms is 0.04 and total RMS is 00.6 thats not too bad is it?

[steppenwolf]

Quote

Point taken, but that's a lot of drift, surely?

A crazy amount of drift, I agree - a puzzling issue but I can't currently see how this is anything other than uncorrected hot pixels and the lines each have the same shape?

[steppenwolf]

Drift can have many causes, not just poor polar alignment. Differential flexure will give rise to 'drift' as will loose components in the optical path. The 'drift' in this case is, however, pretty severe!

[brrttpaul]

Just thinking about the replies and somethings staring me in the face, the erratic lines are always the same direction top left to bottom right, surely if it was drift then it would be different directions on different nights pointing at different targets?

[steppenwolf]

I really think that it is time for some empirical testing here. Perhaps 10 x 5 minute subs in all four quadrants of your night sky without changing the orientation of any components (except the pointing of the mount, or course!) then compare the results having logged carefully where each image set was taken?

[brrttpaul]

Im on a fixed pier and can only see SE to NNW so limited with that, Im going to take lights then process them , repeat with darks then with bias and so on see if those lines come in at a particular source