OIII Narrowband Flats Issue - ASI1600MM-C

rockstarbill

Sam,

Need some help troubleshooting an issue with my ASI 1600MM-C. I have the 8 position EFW and a set of 1.25" AstroDon Gen2 E-Series LRGB and 5nm Ha, OIII, SII. When I take flats, the images look perfectly fine for LRGB, Ha and SII. For the OIII filter, I have a very strange ring pattern on them, and the overall flat doesn't look right at all, in terms of its illumination across the field.

Attached is an example of the problem.

I have discussed this with Don Goldman (AstroDon) and we went through a series of troubleshooting to isolate this problem to the camera. The key step we did was to rotate the filter, and take new flats. This resulted in no change to the pattern in the flat at all.

In terms of a light source, I have used: daylight, laptops, tablets, a flats panel, an evenly illuminated white wall, and a flat panel hanging on the wall with the scope pointed at it. All tests were done cooled and uncooled. All methods yielded the exact same result (that you see attached here).

In terms of scopes I tested this on, I tested a Takahashi FSQ106-ED, TOA-130F, and an AstroTech 65EDQ. All of the scopes showed the exact same pattern in the OIII flats. I tested this with and without a Celestron OAG and a ZWO OAG being attached, and this had no effect on the result at all.

To further this issue with OIII and the ASI 1600, another user (whom I believe is planning to contact you as well) posted a similar issue with the ZWO OIII filters and Baader OIII filters.

Here is my thread on the issue:

http://www.cloudynights.com/topic/568611-strange-oiii-master-flat/

Here is his thread on the issue:

http://www.cloudynights.com/topic/568839-is-something-wrong-with-my-asi1600mm-c-sensor-or-ar-window/

Finally, I wanted to ensure I included additional information on which platforms and drivers I have reproduced this issue on:

ASCOM - Version 6.3, did not test 6.2.

ZWO ASCOM Driver - .23 and .25 were both tested and both showed the issue.

ZWO Camera Driver - 2.1.3 was tested, previous versions were not tested.

Windows Software - Windows 10 Pro, Sequence Generator Pro 2.6.0.17

Linux Software - INDI Server 1.41 on Ubuntu Linux 16.04 LTS, client software was KStars 2.7.5

In all of this testing and in closing I want to be perfectly clear, that the end result did not change at all. In all test cases the flat looked like the attached flat. No difference was found on Windows vs Linux or in driver versions differences, or any of that.

Before you ask -- I do not know if the issue calibrates out when I go to process OIII data because it has been cloudy since I received the filter. Regardless of that, this is a legitimate issue that would be good to get the camera manufacturers perspective on.

All of the actual FIT files for the Master Flats are located here if you would like to download them and examine them yourself.

https://onedrive.live.com/?id=9BD07ECBFEA191DA%2152378&cid=9BD07ECBFEA191DA

I have all of the individual flats as well, so if you need some of those for some reason, let me know that.

Thanks and let me know if you need any more details.

-Bill

rockstarbill

Update:

The current theory to quote Don Goldman:

omething in the A/R coating of the cover plate on the sensor or chamber window is reflecting blue-green (OIII) light. I imagine if you really stretched your blue or green flats, you'd faintly see the same thing. The NB filter just eliminates most other wavelengths and enhances the effect.

I had another CN user do some testing with a 3nm AstroDon filter, and the ASI1600 camera. He uses 36mm unmounted filters, and while his pattern of lines was different than mine, the same general issue existed. I checked my blue and green flats under boosted stretch and could faintly see the same issue, just as Don had suggested.

In the sample light provided by the CN user, the pattern in his flats was not present at all. This bodes well, although raises the issue of how to process flats against the OIII data, in the event that is necessary. The advice I got from Don was:

You could clean your filters to minimize any dust doughnuts and then use the green flat for the OIII filter to eliminate the vignetting and then use Photoshop to clean the few dust doughnuts that may remain.

With that said, I think this firmly places the issue in the ZWO court to determine what about the sensor or window is causing reflections of blue-green light.

bvalente

seems like you should see if it calibrates out before raising the flag higher.

It is concerning, there does seem to be some unusual interference.

if it all cancels out in the end, then maybe no harm no foul?

rockstarbill

bvalente wrote:
seems like you should see if it calibrates out before raising the flag higher.

It is concerning, there does seem to be some unusual interference.

if it all cancels out in the end, then maybe no harm no foul?

B

I am waiting on the weather to test that, but the likelihood of this processing out of the data is slim. In fact, it is highly possible that this aberration will not even show up in the lights. The issue then is that the flat field images are useless for calibrating out other issues (like vignetting).