Baader Clear Focusing filter

L

lvigano

Member
Hi,

I just try to shoot M81 with the Baader Clear filter and some strange it's happening: the image seems to be "saturated". Below some details:
  • Atik One 6.0
  • Sharpstar 94 EDPH (414mm@f/4.4)
  • Capture sw NINA
  • Exp time: 120''
Below the image with CF@120'' with STF:
SNAPSHOT_Clear_0000_120.00_-9.78.jpg

The image statistics:

SNAPSHOT_Clear_0000_120_00_9_78
K
count (%) 99.97014
count (px) 6043246
mean 35825.4
median 35818.0
avgDev 444.1
MAD 358.0
minimum 33158.0
maximum 65520.0

Below the same image with some HT:
SNAPSHOT_Clear_0000_120.00_-9.78_HT.jpg


Now the interesting part: another image @120'' without any filter (STF only):
LIGHT_L_0006_120.00_-10.03.jpg

The image statistics:

LIGHT_L_0006_120_00_10_03
K
count (%) 99.97598
count (px) 6043599
mean 28716.0
median 28710.0
avgDev 387.5
MAD 310.0
minimum 26374.0
maximum 65495.0

I'm running out of ideas ... maybe someone has already got this "problem".

Here is the link with the raw images (also include a 90'' image with the CF which seems "normal"): https://drive.google.com/drive/folders/1ibTIBqpJQn8QWCBqPJnK_5OSEP0m3rIt?usp=sharing

Any idea ?

Best,
L.
 
do you have the transmission graph for this filter? i think Clear is like a Lum filter with no UVIR cutoffs?

you say "without any filter" in the 2nd image but the file is named "L". is there really no filter in the path or is there an L filter?

rob
 
If you are indeed comparing a Baader "C" (clear glass) filter (no cutoff) with a Baader "L" (UV/IR cutoff) filter:
1642030824468.png

then this could be the expected result (with the additional U/V and I/R signal added with the C filter (your camera will almost certainly have no built-in cutoff filters).
 
here on the Baader site:

Baader Clearglass Filter (C) for focusing / dust protection

Baader Premium grade impermeable RGB-C Clearglass Filters.The Baader clear glass filters are used to focus a camera when using very narrow band line filters (due to same mechanical thickness).Alternative function: e.g. our Protective Canon EOS T-Ring uses
www.baader-planetarium.com www.baader-planetarium.com

You have to choose a size to see the graph ...

Yep, I named it "L" in NINA but it was empty.

L.
 
Both frames are heavily overexposed. The "C" frame was taken at 19:38 local time, while the "L" frame was taken three days earlier at 23:28 local time. The relatively modest difference in brightness can be simply explained by the difference in sky background level.
 
not sure what you mean with overexposed. could you kindly check the file "SNAPSHOT_Clear_0000_90.00_-11.07" inside the google link? it was done 2 minutes before the 120'' "C" frame ... the difference in this case is due to the exposure time ?
 
lvigano said:
the difference in this case is due to the exposure time ?
Click to expand...
Exactly! Let's look at the histograms of the two images:
1642035801168.png
1642035844535.png

The only difference is that the 90s "background" peak is just below the 50% signal level, while the 120s peak is just above 50%. All the signal level to the left of the peaks is wasted (i.e. there are no pixels at this level). A correct exposure will have the left edge of the background peak at approximately 0 signal level. anything higher is overexposed (i.e. wasting valuable signal dynamic range).
 
Hi @pfile and @fredvanner : ok, I got your point. This evening I'll try another capture with the C filter @2330 so I can compare "apple" with "apple".

I know that I can have some problems due to the not so dark sky (I know that I cannot compare my pictures with others pictures taken in Namibia :D ), I'm taking picture from here:

1642067332284.png


(https://www.lightpollutionmap.info/#zoom=10.00&lat=45.5164&lon=9.3598&layers=B0FFFFFFFTFFFFFFFFFF)

That's way I switched to mono camera with NB and RGB. I was just curios about the capabilities of the CCD sensor withou any filter :)

For example some statistic about filtered exposures:

LIGHT_B_0003_120_00_10_99
count (%) 99.99583
count (px) 6044799
mean 5188.0
median 5178.0
avgDev 123.9
MAD 97.0
minimum 4448.0
maximum 65534.0

LIGHT_G_0002_120_00_10_75
count (%) 99.99494
count (px) 6044745
mean 8080.5
median 8070.0
avgDev 170.8
MAD 136.0
minimum 7074.0
maximum 65462.0

LIGHT_R_0002_120_00_10_75
count (%) 99.99532
count (px) 6044768
mean 6590.1
median 6577.0
avgDev 147.2
MAD 114.0
minimum 5714.0
maximum 65498.0

LIGHT_S_0000_90_00_11_07
count (%) 99.99992
count (px) 6045046
mean 555.2
median 552.0
avgDev 31.1
MAD 24.0
minimum 376.0
maximum 62106.0

Or other numbers gather from Nina (for LIGHT_R_0002_120_00_10_75) :

1642069255475.png



Maybe there is a simpliest way to get the "optimal" exposure time beside the trial and error :)

I have just done a very quick integration of the L images (@90''):

_test_L.jpg


and RGB images (@120'' - yes this one was not flat corrected 'cause I just finish to gather the flat frames ...):

_test_RGB.jpg


And I'm quite happy with that (I'm not going to submit anything to APOD :D )

(images are added to the gdrive folder, so if you have any suggestion you're more than welcome)

Best,
L.
 
lvigano said:
Maybe there is a simpliest way to get the "optimal" exposure time beside the trial and error
Click to expand...
If you assume your camera response is more or less linear, you can scale your 120s image to move the peak close to the "left edge". I would suggest trial exposures of 1s and 2s. Does this seem very low? Remember, this is a very sensitive camera. A narrow-band filter will pass only about 1% of the light passed by the C "filter", so may require exposures at least 100x as long.
Of course, the resulting images will be virtually black without STF - but that is exactly what you should expect.
 
Yep, I agree with you. I think that the 90'' frames are quite recoverable with HT. In the meantime I had the chance to get some pictures (loaded in google drive):

File_Full_NameClipping_LowClipping_HighMean_0Median_0AvgDev_0MAD_0Min_0Max_0
SNAPSHOT_Clear_0000_1.00_-10.03.xisf065535517516272335214908
SNAPSHOT_Clear_0000_2.00_-10.03.xisf065535706704362946229306
SNAPSHOT_Clear_0000_10.00_-10.35.xisf065535218221787560170965497
SNAPSHOT_Clear_0000_15.00_-10.35.xisf065535301330069273239064954
SNAPSHOT_Clear_0000_30.00_-10.27.xisf06553557735762133104493265523
SNAPSHOT_Clear_0000_60.00_-10.91.xisf0655351112211104199153997365444
SNAPSHOT_Clear_0000_90.00_-10.59.xisf06553516434164122551951488765493
SNAPSHOT_Clear_0000_120.00_-10.91.xisf06553521701216803062351993365518

What do you think ?

I also check the BIAS value which is 329.4 ADU, so from the read noise point of view I can go with 2'' exposure at minimum (just to be sure).

Maybe I have to find a number which is more or less the same as per RGB and another one near the value of the NB filters.
 
lvigano said:
Maybe I have to find a number which is more or less the same as per RGB and another one near the value of the NB filters.
Click to expand...
Yes; a bit of trial and error, but a fairly safe start would be 6s for broadband RGB (about 1/3 of the "clear" bandwidth), 200s for a 7nm NB filter (about 1/100 of the "clear" bandwidth). If you can find darker skies, you can of course try longer exposures.
 
You must log in or register to reply here.
Back
Top