15th Feb 2024: Astro Pixel Processor 2.0.0-beta29 released - macOS native File Chooser, macOS CMD-Q fixed, read-only Fits on network fixed and other bug fixes
7th December 2023: added payment option Alipay to purchase Astro Pixel Processor from China, Hong Kong, Macau, Taiwan, Korea, Japan and other countries where Alipay is used.
Hi I have been trying for a week to understand why my images have so much vignetting but it seem my flat frame are not being applied but i don't know why.
I have a reducer fitted so there is some light loss on the edges but it seem it not getting corrected for in my final images. Flats are taken with a light panel around 30000 ADU tried 20000 and 50000 ADU also.
Here are some images of my flat, light and final integration of just luma to keep it easy but same results on RGB but a lot less on HA,S,O.
I have placed some files on my website link below if anyone wants to run them.
I did a test of not loading flats and dark flats and resulted image is the same so strange.
Now i know my flats,dark flats and dark are all taken at bin1 on the camera and my lights are bin2 but this was just a test i get the same results using bin1 lights.
Process i use is load lights, Dark master, dark flats, flats and then head to tab 6 to integrate i get no errors seems to run all ok.
link to files
https://www.moviecells.net/astro/
Any help would be great not sure what to try next
Mike
Thanks for sharing the data, I'll have a look at that later. It does seem like the vignetting is a bit extreme as if it is clipping. So I'm wondering if something is up with the calibration of your flats for instance.
Thanks that would be great tried a few different ways to take flats they seem to give very close the same results. just to fill in some more info about the setup see below
Scope is a meade 14 inch ACF
Camera Zwo 2600mm
Reducer is a Optec Lepus
Filters 2 in Choma
The reducer should nearly cover the APS sensor size and filters are 2 inch so no problem there. Only thing of note is the reducer is some way from the back of the scope as i have a moonlite focuser before it.
Mike
Mm, I have some trouble downloading the data, they are represented as directories. Is it possible to get at it via ftp or another way to more easily download the entire directories?
Hi Let me zip it in to one file and upload again
Mike
There should be a zip file there now just download and unzip that fora ll the files
Mike
Thanks Mike! Downloading now, I'll probably come back to it tomorrow.
Hi Mike, apologies for the delay, busy weekend. I'll have a look today and report back.
Ok, so a couple of things I'm noticing; you're using a bit of an old APP version (1.079). I would advise to go to at least 1.082 as quite a few things were fixed in between. Also, you're exposing for about 0.25 seconds, that's usually a bit on the low side for flats, advised is to go for about 1-2 seconds and maybe shifting the histogram more to the right as well, it's now at about or just below 50% and given the vignetting I think it's good to move that to 60-70%. Just make sure it's not clipping.
Apart from that, it seems APP is not applying the calibration data at all, so I'm looking into incompatibility somewhere between the calibration data and the lights.
Ah found it, you're binning at 2x2 for the lights and 1x1 for the calibration data. It's not possible to combine that.
Hi I did put that point in the details and i tried bin 1x1 for lights and the same problem happens. I will upload some new files with bin 1x1 lights for you totry.
Mike
Oops, sorry I forgot you mentioned that. Yes, if you provide the bin1 as well that should make for a better test.
Hi just uploading a zip called bin1x1.zip can you try that it has matching lights and cal all bin1x1. Also i did try moving the flats up to about 60% also tried putting a ND filter over the light panel to get the LRGB to the 2 second range and that made no change to final images also.
Mike
Thanks!
So at bin 1 they get applied normally; this is one frame without flats added:
Switching to "l-calibrated" on top of the view window gives me this:
That seemed to work and I integrated everything, with this result:
I then applied LPC correction, with this result:
That is showing a ring effect, which is very likely a light source, causing internal reflections.
Hi Thanks for trying but why is the integarated one look so over corrected just on the right side ? and under corrected on the left
Mike
That is what is left over. The flats correct the vignetting only, light pollution is then what you're left with. It looks like there was a light source from the lower right coming into the telescope.
@moviecells Out of sheer curiosity, what version of APP are you using? If prior to 1.083-beta2, does it help if you enable 32 bit calibration files in tab 2?
To me it looks like the ring is caused by reflections off the secondary mirror holder. What light source do you use for the flats?
Hi I use a flat panel it a A2 drawing light panel and i have a extra sheet of diffuser over the top seem to work well and sit right on top of the telescope with the telescope pointing up.
This is meade SCT not a Newtonian so not sure if the bit about the secondary mirror holder could be true, i am running 1.079.
Do you think the problem a reflection in the flats or as Vincent put more of a light pollution issue.
Mike
Hi How do i download the later versions if i go in to my download in my account it only lists Astro Pixel Processor 1.082 Windows 64bit
Can't see any other way to download
Mike
Here is an overview of all download links: https://www.astropixelprocessor.com/community/appreleases/downloads/
Hi Can you explain this bit
Switching to "l-calibrated" on top of the view window gives me this:
Is there a way to view a calibrated light before its processed ?
Mike
@moviecells Go to the top of the forum page and click on the DOWNLOADS Are Available HERE link.
@moviecells Sure, you can find those options on top of the preview screen, here;
That pulldown menu has various options to show you the result of a process in APP. So when you go to tab 2, calibrate and calibrate your data there, after it has finished you can load 1 sub in the viewer, switch to "l-calibrated" from that top pulldown menu and you can see that one sub as a calibrated sub. If you already have masters for calibration, those will work immediately like this. The advantage of this is that you can check if your calibration (and registering etc) is working as expected before you integrate everything, saving you a lot of time if something is wrong.
Hi Got around to taking some Bin1x1 data last night and flat correction looks fine but i am seeing a strange line pattern down the image see attached tiff file of just blue but i seems to be in all channels can't see any problem in the light files. Let me know if i need to upload the fit files but thought i would ask first.
Mike
Let me try adding file as a png as it seem the tiff file is not displaying correctly.
@moviecells The beta versions are available via the DOWNLOADS are available HERE! link at the top of each forum page.
Hi Yes found that last week thanks the problem now is more to do with the pattern in my processed files any idea what i could be
Mike
@moviecells Sorry, I missed the comments on page 2 of the conversation. Do you see the line pattern as well if you integrate without darks?
I will run a test now, its strange because this is a ZWO ASI294MM that night i used it in its normal Bin2x2 mode matching bin2x2 darks, flats and darkfalts. Last night i used the same camera and set up but in bin1x1 matching bin1x1 darks,flats and darkfalts and can't see the problem. Could it be some kind of pixel pattern due to bining i did not run any dithering on both nights.
Mike