For quite some time I've been messing with my imaging setup in an attempt to get nice round stars and a flat field across the image. Last night I finally had a breakthrough where it all came together. The final piece of the puzzle was the spacing between the flattener and the camera. For months I've had it pretty much bang on the 55mm as per the manufacturers recommendations. In a passing conversation with Troy (who has the same flattener) he mentioned that his performed better at 65mm so last night I gave it a try at 65mm. after a single 10sec frame I could see I had gone too far so I backed it off to 63 and finally settled on 61mm. Here's the comparison of the 55mm (manufacturers spec) and 61mm.
I'm very happy to finally have this sorted!
The moral of the story is don't assume the publicised spacing is the best for your imaging train.
I did plug your photos into CCDInspector. Ideally you need to use a star field and not have a galaxy centered, but the results show that you still have some curvature and you also might have some alignment issues.
If you don't have CCDInspector it's worth a download and try out of free trial. I've found it to be an indespensible tool. With this tool you can quantify your experimentation.
I do have CCDIS however I find it's KPI's can be misleading. Here are the curvature plots for the two images. This would have you believe the 55mm spacing is flatter and have less tilt than the 61mm however visually the 61mm is clearly flatter.
I must try this on a star field to see if the results tell a different story. In any case the elongated stars in the bottom two corners look a lot better to my eye!
I found CCDInspector can give highly variable results, e.g. a series of subs of the same star field with the same filter will show big variations in curvature and tilt. The PixInsight FWHMEccentricity script also shows that individual subs vary widely. Maybe really good seeing is needed to get consistent (and useful) results? In the absence of that then an eyeball probably does a better job.
Great news Peter, who would'a thought ...and yes ultimately it is the eye that counts in the end, not a piece of software that is known to produce misleading results
I found CCDInspector can give highly variable results, e.g. a series of subs of the same star field with the same filter will show big variations in curvature and tilt. The PixInsight FWHMEccentricity script also shows that individual subs vary widely. Maybe really good seeing is needed to get consistent (and useful) results? In the absence of that then an eyeball probably does a better job.
Cheers,
Rick.
Quote:
Originally Posted by strongmanmike
Great news Peter, who would'a thought ...and yes ultimately it is the eye that counts in the end, not a piece of software that is known to produce misleading results
Mike
Cheers Rick and Mike. It's been a long time in the making - quite a relief the camera was not the issue.
Correct Troy - This was the Orion flattener with the Meade ED80. On the Newt I'm using a Parracor with the Televue IS spacer system. For that I found the combination of the 0.375" + 0.5" = 0.875" (22.2mm) spacing did the best job.