Does this master flat look "normal" ?

[DanH]

Hello all,

Sorry if this questions seems dumb, but I wonder if my new CCD (SXVR H18) isn't defective ?
I could be the culprit (at least I hope so) as I am a real beginner in astrophotography.

On my master flat, assembled from 19 flats done with a commercial flat screen, I see many horizontal stripes.

I have uploaded the master flat (32MB) on the endor.uv.es server for those of you kind enough to have a look at the file.

http://endor.uv.es//data/public/2b39d9ae5751751fd99babae0f8f530d.php?lang=en

Thank you in advance.

Dan

[Geoff]

Hi Dan
I have never seen any of my flats look like this one.  Unfortunately I'm not sure what to suggest.
Geoff

[DanH]

Hi Geoff,

That's exactly what I suspected.
I haven't seen any flat looking like this one either...

That is why I'm wondering if the CCD isn't defective.

Dan

[Geoff]

Are your subs normal? If so, then it's unlikely that the ccd is defective. More likely something going wrong in the acquisition of the flats.

[DanH]

Sorry Geoff, I don't understand what you mean by "subs" ? (I'm french speaking...)
I find that Bias and Darks are a bit noisy... not sure if this is normal with the Kodak 8300.

I guess, as the SXVR-H18 is the first CCD from Starlight Xpress with a shutter, that the acquisition from MaximDL is a bit too fast.
For bias, the speed isn't adjustable under MaximDL when using the "autosave" dialog, and maybe the shutter isn't moving fast enough.

I could upload a single bias, dark and flat to endor.uv.es ...  if this is what you mean by "subs".

Thanks for your replies anyway

[DanH]

Just in case, for a short time...

Single bias : http://endor.uv.es//data/public/97a92d710718bbc569eed3a14f7cc9f4.php?lang=en
Single flat : http://endor.uv.es//data/public/46cbe4d7efa893fd8be188f998fa8558.php?lang=en
Single dark : http://endor.uv.es//data/public/44fc3596531bbd70aab8ff6395d24671.php?lang=en

[Geoff]

Hi Dan
Each exposure you take of your object is referred to as a "sub". I think it's an abbreviation for subframe.  So what I'm asking is: do the individual exposures of the object you are imaging look reasonable?

[sreilly]

Best test of a flat is to apply it and see what the result actually is. But after downloading the image and looking at the FITS header I can't tell much. Seems PI strips the exposure data unless this is suppose to be a master. I'd suggest posting a pre-processed raw flat and see what that looks like. What is the exposure length and ADU level of each raw image? I suspect that you should be somewhere between say 30,000-35,000 ADU. What does a light image look like, say through your luminance filter?

-Steve

[DanH]

Hi Geoff,

I have uploaded a single sample of flat, bias and dark. (links above)
The individual subs (thanks for the explanation) look OK, although they are quite noisy for 400 sec exposures.

As I said, this is my first CCD, first use of MaximDL and first use of PI. I was using a Canon color camera before, stacking images with DeepSkyStacker. That was easy !  8)

So there are sooooo many places where I could be wrong !

I have removed my USB hub from the acquisition chain, as I suspect this one as well... I didn't needed so many USB ports with my old Canon. Awaiting better weather to test now !

[sreilly]

No, that single flat doesn't look anything like a normal flat but then your average ADU level is 3882 where it should be more like 30,000-35,000. Use MaxIm's "Information Window" under the view menu and this will give you information about the image. You need to increase your exposure and/or light source. You also need to make sure you are following proper procedure for creating your flats. See this link for a sky flats plug-in for MaxIm. http://winfij.homeip.net/maximdl/skyflats.html John has all his plug-ins on this page http://winfij.homeip.net/maximdl.html

-Steve

[DanH]

Hi Steve,

Thank you for the link !
I was suspecting this master flat from the very beginning, and got the confirmation from yourself.

I'll follow that procedure for flats, hoping it would give me more useful ones.

Did you by chance have a look at the bias ? Is it looking "normal" ?

Thanks again !

Dan

[sreilly]

No Dan, I had just gotten home from a 12 hour shift at the rescue squad and was winding down before going back to bed at 6 am. The dark looks pretty normal and bias frames always look strange to me. The test will be the quality of the image after applying them. Keep in mind that processed master frames in PI will not work in MaxIm as the adu values will be different. In PI they are expressed with values between 0 and 1 while in MaxIm they will be 0-65535 for a 32 bit image. If you are going to do any calibrating in MaxIm, you will need to make either MaxIm master frames (Dark, Bias, Flat) and use those. If calibrating in PI then make the masters in PI and use those.

MaxIm Calibration Frames http://www.youtube.com/watch?v=xJMsyCKcAUU shaky but useful and http://www.cyanogen.com/help/maximdl/HID_PROC_SETCALIB.htm

PI Calibration Frames http://pixinsight.com/tutorials/master-frames/en.html and Harry's video http://www.harrysastroshed.com/Calibration.html

-Steve

[DanH]

Hi Steve,

I realized the difference between MaximDL and PixInsight master frames.
Thank you for pointing that out and refreshing my mind !

And many thanks for your useful comments. As the flats need to be done with the same setting as the set of images, I'll have to wait for a clear sky to check Maxim's plugin.

Best regards,

Dan