Spinning wheel of death: Artifacts from Flats?

[Josh Lake]

http://i.imgur.com/f2XX1.jpg

Has anyone ever see anything like this? A couple of us PI users (addicts) are helping an online acquaintance process some data. His flat frames have never really been quite right, but this is something a bit more severe.

Another strange thing is that his flats don't show the same dust motes and therefore don't get rid of them. This would imply moving dust, but he has a very normal (and automated) setup, I believe. Any thoughts on what might be cause these odd artifacts?

Some flats look fine, BTW, but certain frames and filters have this underlying structure.

Equipment: Planewave CDK 12.5, Apogee U16HC

I'm hoping someone can give us tips on how we might improve his frames and what tests we might be able to run. If you're interested, I could also upload a fair amount of data to the Image Processing Challenge server if you want to try your own hand at it. I can't seem to figure out what the deal is.

[mmirot]

Shutter artifact

Need longer exposures for flats
I need greater than 5sec on my fli

Max

[NKV]

Need longer exposures for flats

+1
With a good uniform light source FlatBox, I see the shutters artifact with 7 seconds of exposure or less.

[mmirot]

I use the same scope, CDK 12.5 but a FLI Proline 16803. 
I use a about six seconds minimum now. These big chip cameras tend to have big slow shutters.

I don't have a problem with moving dust dounuts. Sure a new spot will appears on filter occasionally. I often use a set of flat for a week or more. My scope is pretty well shield from wind in a Dome. I am sure this helps.

I heard the sealing of the apogee filter wheel is not the best.  There may be some tips on the Apogee yahoo group.

Max

[Josh Lake]

Thanks for your help, all... extending the exposure time did the trick as expected.

Can I get your opinion on this new issue? I don't believe my colleague is adjusting the camera at all, but I've noticed a moving vignetting effect between frames. Take a look:

http://i.imgur.com/OdeA0.gif

Is this just the nature of the setup? Can anything be done to keep it more uniform?

Also: Does it matter?! If I get enough flats, even with this shifting pattern, will ABE and DBE be able to deal with all the leftovers?

[bitli]

Thanks for sharing,
I Blinked my flats, and although they 'move' a little after intensity stretching, this is not as much as yours.  Are they stretched or original images ? If they are stretched, may be the difference of intensity between the dark and light parts is in reallity very small (especially in percentage) and may not be that important. If this is the original image, there is probably a problem of using a  less than homogeneous illumination, possibly a light not always at the same place.
My (short) experience is that enhancing the flats as much as possible is very important (I often have sky gradient difficult enough for DBE not to add flat issues).
-- bitli

[Josh Lake]

Thanks for the input. These are indeed all stretched frames using the Blink process, so they look much smoother and more uniform when viewed a linear (naturally!). The are sky flats, so the illumination is definitely changing in intensity. I just wasn't sure if the change in gradient symmetry was normal. I have a Flip Flat for use with my own scope, so we don't run into this issue.

[NKV]

I've noticed a moving vignetting effect between frames.

This is SkyFlat problem. FlipFlat or LightBox should solve the problem.

[Josh Lake]

Does it stand to reason that the same kind of vignetting shift could happen on Light frames as well? Or does the long exposure time take care of it?

[NKV]

Does it stand to reason that the same kind of vignetting shift could happen on Light frames as well?

No.

Or does the long exposure time take care of it?

No.

It's sundown / sunrise problem.

[martin_magnan]

Hi

is your colleague using an adaptive optic device that could in some way shift the image in trying to "follow" something ?

Martin

[Josh Lake]

Hi

is your colleague using an adaptive optic device that could in some way shift the image in trying to "follow" something ?

Martin

I don't believe so (we just correspond through email, I've never seen his facility). Also, he turns off tracking when taking flats, so I'm really surprised that there's still a 'moving spotlight' look. I suppose differential gradients due to sunrise / sunset makes some sense, I'm just surprised that it acts so quickly and randomly. I really wonder if it's something else.

[mmirot]

Actually,  there is a best illumination point near zenith with sky flats.

I believe this changes as the sun goes down.

My automation program, CCD Autopilot, slews to this point.  It then exposes near the point and then moves for each flat.   I will have to check the flats to see if there is a big difference in illumation but I have not noticed it on my CDK 12.5 .

Max

[Josh Lake]

Yeah, this guy is using ACP instead of CCDAP and I don't know if there's a good script for doing what you discuss. Thanks for checking, though, I didn't know there was an optimal spot for skyflats.

[mmirot]

I think ACP should be just fine. I am pretty sure it optimizes flat illumination too.
 I may have never noticed the difference with my sky flats.

The net effect should be small. I usually do a little DBE regardless.

Max