[DiscoDuck (Paul)]

I think I understand my halo issue now. Not solved it, but I am confident re what is happening. And I think there is no solution.

I flipped the Ha filter over, and the halo is still there, but a bit weaker and larger. (Images Dolphin Ha single sub orientation 1 and 2 attached).

The original halo corresponded to a distance between the two reflecting surfaces of about 15mm, about the distance from the filter to the sensor. The new one corresponds to a 17mm distance. So what I think is the case is that with the ZWO 1600 having no AR on its sensor glass cover, light is reflecting off there, and back off the filter and back down.

In the first case, the reflective side of the filter was nearer the camera. Hence the 15mm distance. In the second case, the side with the AR on was nearer, so the light was passing through the filter surface facing the camera and then reflecting back from the inside of the telescope side surface of the filter. It is 3mm thick, and with the refractive index, that accounts for about an extra effective 2mm in the light path. Hence the jump from 15mm to 17mm indicated by the halo size increase.

Bottom line ... the reflection is a bit weaker now, but still there. And really an issue caused by the camera NOT the filter. When the new orientation is stacked, the result is Dolphin Ha.jpg (attached). The halo is very prominent - but only I think because there is so little Ha in this area (Dolphin nebula - mainly OIII!) and that star is actually pretty bright (mag 3.8 I recall). Those bits of Ha at bottom left are very faint, and so it is a very stretched image. (The halo you can see is originally fairly weak - in the original sub it is much dimmer than the amp glow).

It is tame enough now to be correctable - sort of - a clumsy attempt is shown in Dolphin Ha fixed.jpg.

As a quick experiment to test it in the worst case scenario, I took 6x5min images of Horsehead + Flame, with Alnitak in the field of view. Previous attempts with the same scope (different camera and filter) had resulted in much stronger (and multiple) reflections, as in Horsehead old.jpg. The experiment from last night is stacked to give Horsehead.jpg. The weaker and single reflection is almost fixable - Horsehead fixed.jpg.

Sorry for the lengthy blurb. Just wanted to clarify that I don't think the Chroma filter has an issue, but the ZWO camera seems to be the root cause. And even then it's not ZWO's fault, I hasten to add. I believe the sensor with glass is supplied as is by the sensor manufacturer. I'm not sure if it's worth any follow up with either manufacturer then. It seems to be an inherent issue with the reflection from non-AR coated glass in the camera that I may just have to live with and try and process away the rare cases when there are stars bright enough to cause the effect.