Jump to content

vignetting of STL-11000 and 3" Wynne 0.95x on ASA 10" telescope


Recommended Posts

This is a transfered topic from the ASA Yahoo Group.

 

Posted By: konihlav  Thu Apr 25, 2013 3:32 pm

Hi gang,

sorry to disturb with stupid questions, but is there anyone who uses ASA 10"
telescope and 3" 0.95x Wynne reducer/corrector and with 36x24mm CCD camera like
e.g. SBIG STL-11000???

I beg you please, could you show a sample flat field frame (my personal e-mail
address is pablo22 at seznam dot cz)? how much vignetting you get? what filters
do you use? (I use 50mm unmounted Astrodons).

From the PDF papers, ASA claims relative illumination at the edge of 44mm
diagonal to be about 25% at F/3.6 for a 250mm mirror.

My problem is that I don't have STL-11000 but MII G3-11000 from Moravian
Instruments and my first test shot of a flat field looks terrible, like:
http://www.astro.cz/galerie/v/uzivatele/Konihlav/misc/vignetting.jpg.html?g2_ima\
geViewsIndex=1

this is obviously not acceptable for me. I hope the issue is in the front window
being too small see e.g.:
http://www.astro.cz/galerie/v/uzivatele/Konihlav/misc/IMG_4779.JPG.html

thank you all for ANY insight.
best regards
Pavel

______________________________________

Posted By:  s2hao3 Thu Apr 25, 2013 4:11 pm

you said the specs were 25% at the edges for a 44mm diagonal for the ASA corrector
 
your flat is showing about 25% of the peak illumination at the corners: your peak is 44K DN and the corners show a bit more than 12K DN... that’s around 25% so how does changing the sensor to a different brand of camera help you?
 
it seems to me that the results specified by ASA are basically what you are getting. Or did I miss something?
 
Also it looks like you might have a guider tap protruding into the illuminated field at the bottom right. Something is blocking some of the light at the bottom so maybe it is the OAG tap?

______________________________________

Posted By: konihlav Thu Apr 25, 2013 4:41 pm

Dear Richard (the sharp "FLI" guy :) ),

thank you for your quick response.
My math is as follows. Based the center of the frame has about 44000 ADU I would
expect to get around 33000 ADU in the corners provided the vignetting is 25%.
But what I get? around 15000 ADU, i.e. vignetting in my case is 66% ! that's
kinda big.

It looks like a similar case to Canon 5D MarkII DSLR (full frame) and fast
optics. With F-stop much faster than F/4 you can see huge vignetting due to the
obstruction in the camera body.

I have already called Moravian to have a look into it. Anyway I STILL WONDER
what a flat field calibration frame looks like from a typical SBIG STL-11000
camera :) My suspect is that the window in G3-11K is too small (would be the
good case as this would be solvable).

thank you!
Pavel

______________________________________

Posted By: s2hao3 Thu Apr 25, 2013 5:19 pm

I think your problem is with the definition of relative illumination. It doesn’t mean the light intensity drops off 25%, instead it means only 25% survives...
 
if ASA specfies relative illumination in the corners of 25% that means that the intensity in the corners is 25% of what it is in the middle.
 
that’s what I see in your plot
 
what about the protrusion in the lower edge on the right side? Is that an OAG tap?

______________________________________

Posted By: konihlav Thu Apr 25, 2013 6:20 pm

Hi Richard,

I appreciate your opinion and time looking into it. Here:
http://www.astrosysteme.at/images/Corr_3Zoll_Wynne.pdf
you can find the relative illumination for a 250mm scope F/3.8(3.6) it shows a
light fall from 1.0 to 0.75 (i.e. some 25% down). That looks cool to me :) and
also my result - the center part looks great.

I have already feedback from Moravian Instruments (and some other astro friends
of mine) and I am 99.999% sure it's the central window of the CCD camera on
blame (front window) see this image:
http://www.astro.cz/galerie/v/uzivatele/Konihlav/misc/IMG_4779.JPG.html?g2_image\
ViewsIndex=1

they told me it is only 40x28mm big (small) and it's in 29mm distance of the CCD
chip surface :) that's is really silly and they admit it because initially
(years back) they made this size as 2" was standard and they did not think of
any 3" accessories :-) seems funny ;D so they will increase the size to 45x31mm
(from 40x28mm) in order to prevent vignetting at F/3.5. They will send me the
front camera cover part as soon as possible :D

and that protrusion in the lower edge on the right side would most probably be
really the OAG prism so I will move it upwards, should be fine then...

many thanks
sincerely
Pavel

______________________________________

Posted By: s2hao3 Thu Apr 25, 2013 10:47 pm

Yep I agree with the conclusions: the wording of the spec was misleading for me and the result was coincidentally showing a 25% remainder.

The line plot in the first slide really was telling had I thought about it a bit more: there was clearly an abrupt change in slope as the limits of the window functioned as an optical stop.

One point worth making is that if there's any serious light flux drop-off, flat fielding can be adversely affected from a final image SNR perspective. In effect what happens is the final result is flat from a line profile perspective but it will be noisier where the signal level was low prior to calibration.

The line profile can be useful,in illustrating what happens in that scenario. Try flat-fielding a flat image and then look at it using the line profile and it will be apparent.

A good example is shown on pages 27-45 of this:
http://www.narrowbandimaging.com/images/ptc_talk_wsp_2009_crisp_final_comments_web.pdf

In particular see page 33, 34, 38, 39, and 43

If you look closely at the line profile in the upper left of page 33 you will see an inflection point around 10k DN where the vignetting really dominates. That inflection is similiar to the ones in your line profile.

This lens is an f/1.8 so even in the non vignetted regions the cosine^4 light roll off is very pronounced.

Comparing to the larger format f/2.4 lens shown on page 38 and later pages, the delta is striking

Once again it goes to show that using only the sweet spot in the center of these larger lenses offers significant and quantifiable advantages

______________________________________

Posted By: konihlav Fri Apr 26, 2013 10:02 am

Yes, I know your article and I agree.

The image after calibration with flat field would be way tooo noisy. This severe
vignetting is not acceptable.

Good thing is that the solution is on the way. Moravian will produce bigger
front camera cover window and ship it to me ASAP :-) so I hope in 2 weeks I may
have a replacement. It's so obvious. G3-11000 is the biggest camera in G3 series
(there are KAF-6303E and KAF-1001E equipped cameras that are obviously smaller
so the problem is at F/3.6 and G3-11K).

I let you know when I get it and make another flat test.

ANYWAY, IS THERE ANYONE WHO COULD SHOW A FLAT FIELD FRAME DONE WITH SBIG
STL-11000 from this setup to show how it looks like at SBIG ???

thanks!
Pavel

______________________________________


Posted By: o.sedan Fri Apr 26, 2013 10:33 am

Hello Pavel,

Here is a flat taken a couple of days ago, with a STL11k, ASA 10N, 0.95 Wyne Reducer, and L Astrodon 50mm filter  :

http://www.astrosandille.fr/images/brutes/flat-023FL.fit

"The image after calibration with flat field would be way tooo noisy. This severe vignetting is not acceptable"
You are right, but I accept it.

Regards,

Olivier

______________________________________

Posted By: daveddm85 Tue Apr 30, 2013 8:12 am

Olivier,

Your flat doesn't look too bad.

I just purchased an FLI PL16803 and Centerline filter wheel. I am waiting on
parts to connect it to my ASA N10 astrograph.

I look forward to seeing how well it performs with this telescope!

Dave

______________________________________

Posted By: adam.jesion Thu Apr 25, 2013 7:28 pm

Hi,

I am absolutely sure that was a sensor window. Here you have my quick photo without any flat frames:
http://astropolis.pl/index.php?app=core&module=attach&section=attach&attach_rel_module=post&attach_id=79535

Taken with ASA 10" f3,6 + FLI PL16803 (much larger then 11K chip).

Adam
 

 

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...