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.
Hello,
I recently purchased an L-Extreme filter (dual 7nm OIII-Ha) and am getting very strange results that just look plain wrong. I am trying to determine if I may have gotten a defective filter, or if there is something going on with APP. When running the same data through Pixinsight's calibration and registration / stacking tools, the issue does not seem resolved. Here is a link to a google drive folder with raw image files, master darks and flats from Sharpcap and some raw flats for making a master flat.
https://drive.google.com/drive/folders/1f_Iys4aZVdsNB6iKurtd8D1EyHVB1F1o?usp=sharing
I am not sure where the issue is, but any help would be appreciated!
Thanks,
Alex
Hi Alex,
By the looks of it you have lights, flats and a master dark. As APP indicates you will either need dark flats or bias frames as well. Apart from that, what issue do you encounter? What do you mean with you are "getting very strange results that just look plain wrong"?
Wouter
I don't ever use Bias frames with my camera since from what I understand they don't do anything for this model. Attached is a screen grab of the output. This is flat and dark calibrated... so something seems wrong
I agree that you don’t need bias with this camera but you do need dark flats, meaning darks with the exposure time of the flats. We use them with our ASI2600 with L-Extreme as well and without them the results are horrible as well.
Interesting. So you think that could be the issue? And the fact that the L-Extreme is so narrow band that the dark flat may actually be necessary at this point?
It is not the narrow bandness of the filter. You NEEC either biar or dark flats. This is also true with our ASI1600 and ASI6200 and they are monochrome. No matter if I use narrowband Ha or OIII or LRGB.
Thank you. I guess I just have gotten excellent results with this OSC camera and never thought I needed either of these. I will give it a shot and will let you know.
Thanks @wvreeven that's indeed likely the issue. Flats need to be corrected with a bias signal. For new cmos sensors and some other ones, this does indeed not work well. But taking bias frames at at least 0.1-1 seconds seems to work everywhere. Flat darks are therefore a good substitute. The flats need this calibration for APP to get a grip on the vignetting signal etc.
@vincent-mod Unfortunately, this did not seem to do anything for me. I can upload my dark-corrected master flat, if that would be helpful. I guess my main question is if the raw images I uploaded are junk (aka filter is defective) or if they are good images and I am junk (aka I don't know what I am doing!).
Thank you all!
Ok then I should indeed have a look at some of these frames.
Please upload some of the relevant subs to our server;
Go to https://upload.astropixelprocessor.com and use upload3 as username and upload3 as password.
Create a directory named “alpaca-flatsissue” and upload in there. Thank you!
@vincent-mod the files have been uploaded. I provided some raw frames, along with the dark ("dark_20_frames..."), dark-flat and master flat (this is called "22_46_22.fits", that would need to be dark-flat corrected). Let me know if you have any questions, but I am starting to think it really is the filter.
Thanks I will download them shortly.
Thanks I will download them shortly.
Thanks!
@vincent-mod any update on this? Wanted to see if I should get my return in motion! Thank you for the help 🙂
Haha, no I'm not that fast unfortunately, lot of work in between. I'll get to it tomorrow.
@vincent-mod hi Vincent, wanted to see if you had a chance to check in on this. Really frustrating issue! Would love to know if it’s me, my capture software or the filter.
thank you!
You're, as they say, on "hot coals". 😉 I get it, I do have other people's data to get through as well though so I always work at it with the oldest data first. But I'm at yours now..
Ok, so this is how the flat itself looks like (stretched);
To me that looks a bit strange regarding the overall background, how exactly do you take a flat (what is your light source and your setup)?
Also, the darkflat is at 0 second exposure. Darkflats need the same exposure as your flats, just dark (same ISO/gain etc). So you may want to try that first as well, it may be that your sensor is not behaving nicely at 0 seconds.
I use a white t-shirt and my iPad (iPad on top of the scope which is on top of a stretched out t-shirt). I capture it with sharpcap, which allows you to capture all the frames at once and then it averages them together into a master.
The darkflat should be 1 second exposures. I used sharpcap again for this, which captured the 1 second exposures and averaged them together. You should be able to see this if you add the master darkflat as a single sub dark flat (APP only "sees" the exposure length when doing it this way)
Ah ok, I'm wondering if an iPad is able to project a real even, proper full spectrum light. But that's not your issue here at the moment.
Ok, I'll have a closer look, thanks for pointing that out.
Ok, so what I'm also seeing is that the darkflat seems like it isn't completely dark (or has some other issue). This is when I stretch it a little:
I have to say I have little experience with flats and these camera's, but I would normally expect something like this;
You mention that you made sharpcap create the master darkflat? If so, you should make the masters in app as well as it might also cause issues otherwise.
I will try to make the dark flat in APP, just taking the same exposure as my flats, correct?
Yes, same everything, just dark. Also, masterdark and other calibration is preferred to be made in APP.
I have had issues with taking flats with an iPad as well. Now I simply point the telescope with the camera attached to a white wall or door and make sure that that gets illuminated by natural light. No artificial lights! The illumination can be increased or decreased by opening or closing doors, curtains etc. Works like a charm. I also place a with t-shirt over the front of the telescope to even more ensure a homogeneous illumination of the sensor.
If you have a permanent setup and cannot remove the telescope then placing a white piece of paper or card board or so could work.
Indeed. Or using a flat-panel, I've done that all the time and it is suitable for that because it has an even distribution of light and has a broad spectrum. When I lived in NZ and didn't have my gear, I managed to take flats by using a white sheet over my camera, pointed at a table with white sheet, which bounced sunlight. This way I could avoid the illumination differences of the clouds in the sky. 🙂
It might actually also be the reason for this when i think about it. First we need to factor out that sharcap is the issue, but if that doesn't change anything I also think it might be the light source.
I re-created the flats using raw flats and raw dark flat files and re-did the integration and got the same result. To me, it really is looking like a data issue. I have uploaded the raw flats and dark flats to the drive.
Thanks, I'll have a look at those tomorrow. It might be worth looking at using something different from an iPad as light source, which will be better in any case. I'm still wondering about the flatdarks signal as well though. How are you making sure it's completely dark?
My iPad was totally fine for flats for a year, so I imagine that isn't it (I use a white PDF and have it open on top). The master dark flat may have been an issue, but the new flat darks I uploaded should be totally dark.
Did you take flats this way with the L-Extreme filter before and got good results?
I never got good results with the L-eXtreme. That is my issue
Even with the new flat method, I am still getting the same image output. So it may be my data?