Sun Ha 20/06/17 Flats experiment

[james_screech]

I thought I'd experiment with flats today. Used cling film over the objective to get flats both full disk and close up, with mixed results. Captured 1000 frames and stacked 500 for each. The images below have just been stacked in AS! and sharpened with Registax wavelets, no other processing.

Full disk with flats

 

Full disk without flats.

 

Close up with flats

 

Close up without flats

 

Close up disk edge with flats

 

Close up disk edge without flats

 

The full disk with flats could be improved with more processing, however the NR looks far worse with flats!

The close up on the disk does improve with flats, however the close up at the solar edge has worse NR with flats.

I know all the images could be improved with more processing, however I wanted to compare just the effects of flats, as you can see the results are mixed. More experiments are called for I think.

 

James

[james_screech]

Now the best of each with a little be more processing.

 

[xtreemchaos]

great shots mate, as for flats i dont use them with my HA shots, i don't get NRs so i feel no need, might have to when i change to a mono cam. well done nice comparison. thanks , charl.

[Macavity]

Nice bit of research - I thoroughly approve! Having been some
of this way myself, I refer you to some of my posts... or not! 
I cannot remember what I put now?!? lol. But from memory:

My general observation was that FULL DISK flats did not really
cure Newton's Rings. I did succeed in alleviating most of them
via "drifting". N.B. that I did not "stop" the scope, but used
EQMod to input custom tracking rates +/- 10-20% nominal.
(This keeps the Sun, "sweet spot" etc. from drifting too far!) 

But eventually, I found that *remembering* to change these
rates was tedious, so I switched from a "CMOS" to "CCD" chip.
The alternative was to try the various TILTING strategies? But
I see claims "no amount of tilting" an ASI120MM works"? 

BUT THEN my CCD chip proved to have non-uniform pixels,
so I am still using FLATS for 2x2 matrices / higher! I am able
to "get away with it" re. full disks by *rotating* the sun to a
correct orientation - It smooths out the pixel edges a tad! 

I did learn (Thanks SGL Solarites!) that to use flats effectively
for solar "quadrants", I do need to *re-center* the sun! My
2x2 Flats then *just about* manage to cover the full chip area.
Part of the problem is that processing software gets "fooled"
by large areas of BLACK around the sun! "PIPP" might helps? 

I would still like to see *practical* evidence of 100% effective
use of a full disk flat! [part teasing] I still revisit this from time
to time... Supermarket "bag" flats, multi-layers of cling film...
Reconstructing flattened disks using IRIS artificial suns etc.

Perhaps you will find (as I did) that once the Newton's Rings
have been removed the native image looks rather better...  

[Pete Presland]

Some very nice detail visible there, the NR's are mostly cured by a tilt adjuster being fired.