Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Help! I can't remove the effects of the focal reducer from my image

  • Please log in to reply
17 replies to this topic

#1 Tulloch

Tulloch

    Soyuz

  • *****
  • topic starter
  • Posts: 3,818
  • Joined: 02 Mar 2019
  • Loc: Melbourne, Australia

Posted 14 January 2022 - 07:35 PM

Hi there, I recently took some frames of Orion using my 6" SCT, Meade f/6.3 focal reducer and Canon 700D camera. For the most part these turned out pretty well, but I noticed that once I started really ramping up the stretching in DSS and Photoshop, the effects of the focal reducer showed through and I wasn't able to get rid of it.

 

I stacked the best 30/36 frames in DSS, camera settings were ISO 1600 and 20sec exposure time. I took 20 darks at the same ISO/shutter and 20 flats at ISO 1600 using the "white T-shirt" method during the day in the same configuration as the lights. I wasn't sure what shutter speed to use for the flats so I tried 1/500, 1/640, 1/1000 and 1/2000 sec exposures. The flats looked like this.

 

6in SCT focal reducer flats.jpg

 

When I stretched the image of Orion, I got this, where the effects of the reducer are quite visible as a "ring" around the nebula (preliminary image, still needs more work smile.gif).

 

Orion Jan2022 1-2000flats ps1sm stretched.jpg

 

I was hoping that the flats (here I used the 1/2000 sec ones) would remove this brighter ring, am I asking too much of the removal process technique?

 

I tried the other flats with a similar result, is there anything else I can do with this image? Dark flats?

 

Thanks,

 

Andrew



#2 jrschmidt2

jrschmidt2

    Vostok 1

  • *****
  • Posts: 161
  • Joined: 30 Jul 2020
  • Loc: Fitchburg, WI

Posted 14 January 2022 - 07:39 PM

Your flats are likely way too short and underexposed. You should be shooting for a histogram roughly peaked in the middle, but not clipped.

You may also need bias or dark flats to get this to work. I'm not sure what best practice is for a dslr.

Edited by jrschmidt2, 14 January 2022 - 07:40 PM.


#3 Tulloch

Tulloch

    Soyuz

  • *****
  • topic starter
  • Posts: 3,818
  • Joined: 02 Mar 2019
  • Loc: Melbourne, Australia

Posted 14 January 2022 - 07:45 PM

Your flats are likely way too short and underexposed. You should be shooting for a histogram roughly peaked in the middle, but not clipped.

You may also need bias or dark flats to get this to work. I'm not sure what best practice is for a dslr.

OK, thanks. The 1/500 sec flats have a histogram like this, would these be better?

Attached Thumbnails

  • 1-500 sec flat histogram.JPG


#4 BQ Octantis

BQ Octantis

    Fly Me to the Moon

  • *****
  • Posts: 5,584
  • Joined: 29 Apr 2017
  • Loc: Mid-Atlantic, USA

Posted 14 January 2022 - 08:39 PM

G'day Andrew,

 

For my 600D, I've gotten outstanding results with flats at ISO100, set to Aperture Priority mode. This will expose the histogram properly and minimize the noise in the flat.

 

From what I've been told, DSS requires bias frames to divide the flat properly. Use the same ISO as the flats, Manual mode, Ts = 1/4000, aperture covered. (Yes, you could alternatively use flat darks; set Ts to whatever the Ts was for the flats. But the dark current noise at that exposure won't make a perceivable difference.)

 

From what I've seen, if your flats aren't correcting the ring, it's probably incorrect backfocus between the focal reducer and the camera. The backfocus required for a focal reducer has very little wiggle room. The typical required distance is 55mm.

 

Cheers,

 

BQ

 

P.S. Post a light and a flat somewhere and I can see if there's a ring from the focal reducer. I use RawTherapee, which doesn't require a bias—it just subtracts the average bias level recorded in the EXIF header.


Edited by BQ Octantis, 14 January 2022 - 08:52 PM.


#5 Tulloch

Tulloch

    Soyuz

  • *****
  • topic starter
  • Posts: 3,818
  • Joined: 02 Mar 2019
  • Loc: Melbourne, Australia

Posted 14 January 2022 - 09:39 PM

G'day Andrew,

 

For my 600D, I've gotten outstanding results with flats at ISO100, set to Aperture Priority mode. This will expose the histogram properly and minimize the noise in the flat.

 

From what I've been told, DSS requires bias frames to divide the flat properly. Use the same ISO as the flats, Manual mode, Ts = 1/4000, aperture covered. (Yes, you could alternatively use flat darks; set Ts to whatever the Ts was for the flats. But the dark current noise at that exposure won't make a perceivable difference.)

 

From what I've seen, if your flats aren't correcting the ring, it's probably incorrect backfocus between the focal reducer and the camera. The backfocus required for a focal reducer has very little wiggle room. The typical required distance is 55mm.

 

Cheers,

 

BQ

Thanks BQ, I've got some bias frames around here somewhere, I'll try again using those and the 1/500 sec flats.

 

I've had to play around a bit with my setup, and optimised my focal train for the 9.25" (which I didn't bother changing for the 6") - I used a 15mm spacer between the SCT adapter and T-ring which gives me 0.828 arcsec/pixel (from Astrometry.com) for a f/l of 1071mm, or f/7.1 which is a bit off the f/6.3 that I'm supposed to get.

 

 

P.S. Post a light and a flat somewhere and I can see if there's a ring from the focal reducer. I use RawTherapee, which doesn't require a bias—it just subtracts the average bias level recorded in the EXIF header.

OK, did you want the raws?

 

Andrew


Edited by Tulloch, 14 January 2022 - 09:40 PM.


#6 BQ Octantis

BQ Octantis

    Fly Me to the Moon

  • *****
  • Posts: 5,584
  • Joined: 29 Apr 2017
  • Loc: Mid-Atlantic, USA

Posted 14 January 2022 - 10:17 PM

Thanks BQ, I've got some bias frames around here somewhere, I'll try again using those and the 1/500 sec flats.

 

I've had to play around a bit with my setup, and optimised my focal train for the 9.25" (which I didn't bother changing for the 6") - I used a 15mm spacer between the SCT adapter and T-ring which gives me 0.828 arcsec/pixel (from Astrometry.com) for a f/l of 1071mm, or f/7.1 which is a bit off the f/6.3 that I'm supposed to get.

 

 

OK, did you want the raws?

 

Andrew

 

Again, the backfocus (the distance from the glass to the sensor) is typically fairly critical for a focal reducer/field flattener. From what I can tell, it has to do with light cones and its reflections that the designers accommodate. Check the documentation for the correct value.

 

Yes, raws, please. I just feed them into RawTherapee and apply my standard calibration workflow. Just one flat is all that is required for vignette and dust mote removal. If you want to throw in a raw dark, that would be good, too. Just 20 seconds at f/6.3 might be a little challenging to stretch, but it's worth a look…

 

BQ



#7 Tulloch

Tulloch

    Soyuz

  • *****
  • topic starter
  • Posts: 3,818
  • Joined: 02 Mar 2019
  • Loc: Melbourne, Australia

Posted 14 January 2022 - 10:46 PM

I tried again with 1/500 sec flats and 15 bias frames - the ring is still there.

 

Orion Jan2022 1-500flats Bias ps1sm.jpg

 

Yes, raws, please. I just feed them into RawTherapee and apply my standard calibration workflow. Just one flat is all that is required for vignette and dust mote removal. If you want to throw in a raw dark, that would be good, too. Just 20 seconds at f/6.3 might be a little challenging to stretch, but it's worth a look…

OK, here they are - the ring is clearly visible in the light, no need for stretching smile.gif

https://drive.google...2H-?usp=sharing

 

Andrew



#8 BQ Octantis

BQ Octantis

    Fly Me to the Moon

  • *****
  • Posts: 5,584
  • Joined: 29 Apr 2017
  • Loc: Mid-Atlantic, USA

Posted 14 January 2022 - 11:09 PM

There's no doubt that the cause is in the optical train. The flat corrects the vignette perfectly, but the halo remains:

 

gallery_273658_12412_172470.jpg

 

A halo from incorrect backfocus cannot be corrected with a flat.

 

To rule out the sensor, try shooting a light and a flat without the focal reducer. I'm wondering if that will also remove the interesting flare off Hatysa…

 

BQ



#9 Tulloch

Tulloch

    Soyuz

  • *****
  • topic starter
  • Posts: 3,818
  • Joined: 02 Mar 2019
  • Loc: Melbourne, Australia

Posted 15 January 2022 - 12:00 AM

Thanks BQ, that flare is apparently due to internal reflections off the baffle tube ...

 

I'll try again with the correct backfocus on the 6", I needed to reduce it with my 9.25" as it was too long for the Evo mount :)



#10 VincenzoZito

VincenzoZito

    Explorer 1

  • -----
  • Posts: 74
  • Joined: 31 Oct 2019

Posted 15 January 2022 - 10:21 AM

Hi,

Test without reducer.
If the halo disappears it is not the deflector.
In this case use the reducer only for small objects in the central area and crop the image.
No free soup.
Try other reducer later.



#11 Tulloch

Tulloch

    Soyuz

  • *****
  • topic starter
  • Posts: 3,818
  • Joined: 02 Mar 2019
  • Loc: Melbourne, Australia

Posted 15 January 2022 - 09:13 PM

Yep, sound like I need to re-look into my imaging train.

 

Or, a little non-standard cropping to make lemonade out of lemons smile.gif

 

 

Attached Thumbnails

  • Orion Jan2022 1-500flats Bias circle_filtered ps1smcrop.jpg

  • Pelayo and Borodog like this

#12 BQ Octantis

BQ Octantis

    Fly Me to the Moon

  • *****
  • Posts: 5,584
  • Joined: 29 Apr 2017
  • Loc: Mid-Atlantic, USA

Posted 15 January 2022 - 09:27 PM

Ha! Nice!

 

BQ


  • Tulloch likes this

#13 Borodog

Borodog

    Soyuz

  • *****
  • Posts: 3,681
  • Joined: 26 Oct 2020

Posted 15 January 2022 - 09:32 PM

Yep, sound like I need to re-look into my imaging train.

 

Or, a little non-standard cropping to make lemonade out of lemons smile.gif

youcandoit.gif



#14 dcaponeii

dcaponeii

    Vanguard

  • -----
  • Posts: 2,388
  • Joined: 01 Sep 2019
  • Loc: Waxahachie, TX

Posted 16 January 2022 - 08:37 AM

If you're going to do a lot of DSO until the planets return you'll find that AstroPixel Processor's light pollution removal will take care of that halo.  Can't fix the flare but the halo is a piece of cake with APP.


  • Tulloch likes this

#15 WadeH237

WadeH237

    Cosmos

  • *****
  • Posts: 8,020
  • Joined: 24 Feb 2007
  • Loc: Ellensburg, WA

Posted 16 January 2022 - 09:13 AM

Usually, a vignetting ring after flat calibration indicates a light leak somewhere.  If you find and address that, the ring should disappear.


  • jonnybravo0311 likes this

#16 Tulloch

Tulloch

    Soyuz

  • *****
  • topic starter
  • Posts: 3,818
  • Joined: 02 Mar 2019
  • Loc: Melbourne, Australia

Posted 16 January 2022 - 05:34 PM

If you're going to do a lot of DSO until the planets return you'll find that AstroPixel Processor's light pollution removal will take care of that halo.  Can't fix the flare but the halo is a piece of cake with APP.

Thanks Don, how does APP tell the difference between the halo and the other parts of the image with a similar intensity?



#17 Tulloch

Tulloch

    Soyuz

  • *****
  • topic starter
  • Posts: 3,818
  • Joined: 02 Mar 2019
  • Loc: Melbourne, Australia

Posted 16 January 2022 - 05:40 PM

Usually, a vignetting ring after flat calibration indicates a light leak somewhere.  If you find and address that, the ring should disappear.

Hmmm, interesting, thanks for that. 

 

The image was taken in my Bortle 6 backyard with lots of background light (from the surrounding houses etc) using a DSLR. I put the eye-cap over the back of the camera during recording, but it's not exactly a light proof box :).



#18 dcaponeii

dcaponeii

    Vanguard

  • -----
  • Posts: 2,388
  • Joined: 01 Sep 2019
  • Loc: Waxahachie, TX

Posted 16 January 2022 - 07:25 PM

Thanks Don, how does APP tell the difference between the halo and the other parts of the image with a similar intensity?


You place your sampling boxes along the margins inside and outside the edges. I think that tells APP that those areas are supposed to be the same in the light pollution model it creates.
  • Tulloch likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics