Odd OIII filter gradient - Why?

[davew]

To be honest I'd be even happier if it wasn't reflection. If it was then we'd all be on the look out with Astrodon OIII filters. If it's what you suspect then it could be a one off never to be repeated.

I wouldn't say it was an Atik fault because they didn't manufacture the chip and many of the other Sony equipped cameras are the same. On the cameras I have looked at there are very few that shield the chip and even when some do the contacts are still exposed to some degree. Truesense are shielding the actual chip these days by the looks of it.

In a general sense I'd be happy if it was a read defect. For Sara I'm disappointed with her bad luck,

Dave.

[Oddsocks]

One of the reasons I thought it might be a reflection / scatter artefact is because the banding fades in line with the vignetted field and does not reach the full width of the sensor, for a power induced read artefact I would expect the banding to fully cover the sensor but you can see in both the normal and inverted image that the banding fades towards the edges in line with the illuminated field.

I think reflection artefacts may be fairly common with highly reflective interference filters from any manufacturer, I can't see how they can be avoided when gold or other metals are used in the filter deposition layer, but the artefacts are not seen easily without significant stretch, but also if it is a reflection artefact it should be removed with a good flat frame but Sara stated the banding became worse with flat frame removal which is odd.

I don't always trust the processing software to match the correct lights and flats in batch processing and prefer to work on individual groups, I find this to be more reliable so for example I would only (stack) process all OIII flats into a master flat, all bias into a master bias, all darks into a master dark and all OIII lights into a master light, repeat for all the other filters individually and then take the calibrated frames for combination in PS, that way I have full control over each stage and tinker with individual light masters to reduce star bloating before combining into a final RGB. I know it takes much longer but I figured if it took me three or four nights to record the data the least I can do is spend half a day in final processing.

Certainly an interesting problem though, I wish Sara good luck with nailing the fault and look forward to reading what the solution turns out to be.

William.

[swag72]

I always do calibration separately for each pane of an image for example, with matched bias and flats. I don't trust a PC to do it either!!

My interest is that if this is potentially a camera issue, why would it not show equally in each filter? Just because of the reflective properties of the narrowband filters?

[ollypenrice]

I like WIlliam's point about the artefact being vignetted. That is surely more consistent with an optical than an electronic explanation.

Now, the flats make it worse. This suggests that they are over correcting, so any optical effect is stronger under flats-shooting circumstances than imaging circumstanances, and wouldn't we rather expect this if stray light is the cause, the flat panel being so bright?

An improvised mask surely beats a wasted down-time if the issue does turn out to be optical.

Olly

[Earl]

Can you replicate it without any filters been used?

[swag72]

If I am taking the flats at an average of 22k ADU - Are you suggesting that this is too bright and that the gradient wouldn't show if the ADU was lower?

[swag72]

Can you replicate it without any filters been used?

I don't know Earl - I can take a flat (albeit just against a white wall in the obs) with no filter at all. I'll give that a go.

[Earl]

Can you take a set of flats with the white wall to keep the conditions the same, all filters then without, it might show something been able to compare them all.

[swag72]

I can, but at the moment I can't get the 20k ADU figure as the door of the obs is closed to prevent light coming in from the side. The image that I took this morning at 9k showed the gradient coming through.

[Oddsocks]

Hi Sara,

I think the strength of a refection artefact depends on two factors.

1: the wavelength of the light, my reasoning is that if you look at a red object in red light the object will show bright red but if you look at at the same object lit with green light then it appears black, now for narrow band imaging you are essentially isolating different parts of the spectrum and if the artefacts are reflection artefacts from parts of the camera then I would expect those parts to show a different level of reflectance depending how the material reflecting the light reacts to the wavelength of light illuminating it. So differing amount of light reflection depending on the pass band of the filter.

2: I can plainly see from my own filters that the OIII seems "shinier" than the SII and Ha, I don't know why this should be so, but obviously as the filters are made to transmit different pass frequencies then there must be subtle surface differences between them.

William.

[swag72]

I have just taken an image with no scope and no filters. I managed to get it to 33k ADU and it does not show any hint of the banding that I am getting with the scope and filter in place.

[davew]

So. I wonder if this brings us back to the front of the camera or light reflection off an internal anodised surface. Although I would suspect IR leak if that were so.

Also keeping in mind the possibility of a camera read defect.

Dave.

[Gina]

This is very strange.  I also have the Atik 460EX, EFW2 and Astrodon 3nm OIII filter.  I have used these with MN190 (1000mm), ED80 with 0.85 FR/FF (510mm) and 200mm SLR lenses and not had the problem.  I've also used the 3nm OIII with the 314L+ with the scopes and various lenses (without FW).

[FLO]

I really don't think it is reflection from the CCD edge or the CCD chamber (that would be a first) but I haven't seen an electrical/interference pattern like this either. I can understand your reluctance Sara to return the camera to your supplier, there's a lot of miles between Spain and the UK. Perhaps you could try an image without the filter wheel, just a nosepiece and filter, to determine where the problem lies. 

HTH

Edit: I missed your post #36. So it might be a reflection from within your filter wheel? 

[Mark72]

Hi Sara,

What did you image? Could you post uncalibrated sub and may be a flat at 30Adu? Was there Moon around?

Mark

[swag72]

Hi Mark. This has been a recurrent issue with all of my OIII images with the 3nm filter.

Atik and my original retailer are looking into it and I am supplying various flats / images as they request as a process of elimination. So far the first image taken with nothing other than the camera is clean, so the sensor is fine.

[davew]

This sounds more like a reflection issue the more that gets said. I've dumped the idea it may be the contacts on the chip and am going with the full chip reflecting.

I'm intrigued that Atik have fixed a theoretical problem that doesn't exist so is this the problem we now see ? If so would it effect all other makes of filter wheel ? If so are all the other manufacturers aware ?

I've looked back over the whole thread and no one is saying or even intimating that this is a fault with the camera producer. Something is very definitely wrong here so I keenly await the original retailer and Atik's findings on the filter wheel's ability to stop light paths where they're not wanted and possibly bouncing back and forth.

[FLO]

This sounds more like a reflection issue the more that gets said. I've dumped the idea it may be the contacts on the chip and am going with the full chip reflecting.

I'm intrigued that Atik have fixed a theoretical problem that doesn't exist so is this the problem we now see ? If so would it effect all other makes of filter wheel ? If so are all the other manufacturers aware ?

Sara is not using an Atik filter wheel. Of that I am sure

Steve

[davew]

Didn't say she was  The problem Atik fixed in their filter wheels may not be fixed in everyone else's wheels. Are the other manufacturers aware of it ? If there's a chance that reflections are caused by adjacent filters then we could all have a problem with Astrodon OIII filters if we don't use Atik's wheels with this camera.

It may surprise you to hear I was seriously considering an Atik 490. Same size and type of chip. Double  I also use Astrodon filters

Dave

[FLO]

It may surprise you to hear I was seriously considering an Atik 490. Same size and type of chip. Double  I also use Astrodon filters

It wouldn't surprise me at all  I recently chose an Atik 490EX for use with my f3.9 Borg 71FL. I had planned to use a 460Ex but the 490EX is the better choice for short focal lengths. The 71FL will have a filter drawer so I won't need a filter wheel, and I'll be using Baader filters. 

I am confident there are no reflection issues with the EFW2 filter wheel because I know how much time and effort they have put into it. It is a remarkable piece of engineering, milled from solid aluminium! They knew they had to get the EFW2 right because the previous model was a bit pants (sorry Atik) and their next step is to launch cameras with integrated filter wheels. 

Steve

[davew]

I go even shorter on FL than you do. The 490 pixel size was created just for me ! Others can use it too I would use a Filter drawer also.

I've never heard a contrary word about the EFW2 so it looks like if a 460 / 490 is used then the front of the wheel opening requires masking with any other make. If I understand what's been written then that's what Atik have already done with theirs.

I foresee problems with that so we'll see what the other bunch come up with ? A mask can be called a baffle or it can be called a diaphragm. That implies that it could, if not properly installed, lead to chip shading or a reduction in FR.

Dave.

[Sp@ce_d]

Think I may just hold off adding a 3nm to the collection until we find out the root cause then. I agree the EFW2 is a beautiful albeit expensive piece of engineering. One thing that hadn't occurred to me until it came up here was the reduction of the aperture in the 2" nosepiece. Steve, I take it if one moves to 2" filters then this will need changing also?

[davew]

Think I may just hold off adding a 3nm to the collection until we find out the root cause then. I agree the EFW2 is a beautiful albeit expensive piece of engineering. One thing that hadn't occurred to me until it came up here was the reduction of the aperture in the 2" nosepiece. Steve, I take it if one moves to 2" filters then this will need changing also?

If I've read it correctly then it appears like a 1.25" nose piece adaptation not a whacking great chunk of metal that it could look like I've said.

[Sp@ce_d]

If I've read it correctly then it appears like a 1.25" nose piece adaptation not a whacking great chunk of metal that it could look like I've said.

It's been a while since I dismantled my setup but from what I remember there's a reducer in the back of the 2" nosepiece thats not removable. I didn't pay much attention as I use 1.25" but thinking about it now it'll vignette 2" filters & a larger sensor. I'm wondering if you need to get a different nosepiece when you get the 2" filter wheel? I hadn't seen it mentioned before.

[FLO]

It isn't a heavy clunky thing, it looks just like any other 2" nosepiece but instead of a 2" aperture at the rear it is 1.25".  

I am at home now but I'll open an EFW2 and photograph the nosepiece Monday. 

HTH, 

Steve