Weird star shape with Ha filter

[luka]

Hi

I got a 2nd hand Orion extra narrowband imaging Ha filter and I am getting some really weird shapes of stars. See, for example, the attached crop photos for 120s and 30s exposures + single 120s uncropped exposure of M42 @ ISO1600. This is when focused as good as it gets on a brightest star using live-view on a laptop.

The same happens with two DSLRs models, Canon 450D and 1200D, both fully IR modded and not debayered (I am aware of the disadvantages).


Stars change shape across the image.
When slightly defocused the brightest star has a crescent moon shape and a complicated dual-lobe shape when slightly defocused on the other side of the focus.
Visually the filter looks perfect.
Combining the Ha filter with Orion luminance IR-cut filter gives the same result so the Ha filter is not letting through the FIR.
Stars look round without any filters or even with a Baader SII filter which has a relatively similar wavelength.
Field flattener does not seem to make a difference (the attached images are taken with the flattener).

Scope is ED80. The filter was sticky-taped to the front of the flattener - would tilting of the filter cause this? I know that sticky-tape is not the best but the filter looked very well placed and attached and also this was just a test, not a permanent solution.

Any ideas?
I saw the same effect with filter taken on/off multiple times over several nights.

Thanks in advance
Luka

edit: Corrected filter name, it is "extra narrowband" and not "ultra narrowband"

[glend]

Just how narrow is Ultra Narrow? A quick internet search suggest the Orion Ultrablock Narrowband filter is for visual use, but the may not preclude imaging through it. I could not find a spectrum pass chart for it. There is definitely some tilt or optical aberation.

[multiweb]

Could be a collimation or tilt issue. Tighter stars with a HA filter will exacerbate the problem.

[billdan]

It looks pretty bad, not sure what the issue is, when filters age they de-laminate but it should be obvious to the eye if that is so.

I don't think it is the Orion Ultrablock, according to this chart it does not pass Ha.

Bill

[luka]

Thank you for your replies :thumbsup:

Glen, the filter ring says "Orion extra narrowband imaging filter" so I assume it is an imaging filter. Also apologies, I used word "ultra narrowband" instead of "extra narrowband" in my original post. It has been corrected now in the original post.
The seller said that the filter is, and I am quoting:
"12nm?"
Note the question mark at the end so I don't think they were sure. Internet suggests that "extra narrowband" is 7nm.

Marc, I was not aware that Ha (narrowband) will exacerbate the tilt/collimation problems. I suspect that there is tilt in my system but it never bothered me before. I am using a refractor so collimation is not an issue.

I have found this page (http://www.narrowbandimaging.com/ccc18_bubble_nebula_in_red_page.htm ) which shows how collimation can affect the focus. If tilt has similar effect, it can explain the crescent shape and the double-lobe shape when out of focus.

I will try playing with this when it stops raining and also try the SII filter again which I thought worked OK. I don't trust my memory from events after 2am :lol:

[multiweb]

Sorry, what I meant is that because you get tighter stars in HA you tend to see bad star shapes better. When they're bloated and blurrier the seagull shape can get smeared so it's not as obvious off axis. Tilt can definitely do this so squaring the camera is important. To diagnose you could try to rotate the camera in increments of 90 degrees and do a short exposure and see if the flaring follows or not. Could be the sensor or the whole camera in the drawtube.

[luka]

Hi Bill, you posted your reply while I was typing so I did not see it :)

It is "Orion extra narrowband imaging filter", not ultrablock. My bad, apologies again for starting people on the wrong track.

No delamination is visible, the filter looks OK.

I am suspecting tilt now, have to do a bit of reading how to quantify it and also about the Skywatcher ED80 Crayford focuser and its week points.

[billdan]

It has to be the method of mounting Luka, sticky-tape method must be introducing tilt. As you say with all other filters mounted the stars look good.

[Slawomir]

I agree with Biil. Using sticky tape to hold things up in the optical train, in particular when imaging, sounds like a very bad idea to me.

[DJScotty]

Hi luka. This is my old filter if I am not mistaken.
I had exactly the same issue.
You have to have this filter as close as you can to the sensor.
Plus I agree with the other responders sticky tape isn't great.
A filter wheel is a must
Cheers
Scott

[luka]

Hi everybody, thanks for your replies. I was planing to do more testing but I absolutely didn't get any time to do it yet. Hopefully this weekend and I will update this thread with the results of my findings.

Scott, yes, this is your old filter. Did the other filters I got from you have the same problem?
Thanks for the hint about the distance. I have it mounted very far from the sensor, probably as far as 100mm, in front of the flattener. I need to 3D print an adapter to mount it inside my filter mount, the sticky taping at the flattener was just a quick fix... but after your comments the sticky tape was clearly not the cause for the star shape.

Actually, what would be the reason that the filter distance to sensor is affecting the star shape?

[DJScotty]

None of the other filters had this effect.
My discussions with a responding member at the time of my difficulties with the filter led to reflection.
Putting the filter in a filter wheel as close as possible to the sensor fixed the problem.
Did you not also get a filter wheel? Obviously there is a reason why you are not using it. ..

[luka]

Scott, few of us are building astro cameras (http://www.iceinspace.com.au/forum/showthread.php?t=146493) but until that is finished I am just playing around with my DSLR and the filters. The reason I cannot use the filter wheel is that my flattener requires 55mm distance to the sensor while Canon DSLRs have 44mm distance of the sensor to flange. The filter wheel is thicker than 11mm so that rules it out (or the flattener).

I do have a DSLR filter holder but the rings on these filters are too thick and don't fit. I will 3D print a holder but didn't get around to doing it yet. This will bring the distance of the filter to sensor to about 50mm, hopefully this will solve the problem or at least help a bit.

[DJScotty]

Fair enough. I knew there'd be a good reason.
The baader filter threads are the standard threads (m48 x 0.75?) Whereas the Orion threads have a 0.7 pitch. It might be the other way around.
But anyway good luck with it

[luka]

Just a quick update, I 3D-printed a mount for the filter glass (without ring) and placed it into the holder (http://www.teleskop-express.de/shop/product_info.php/info/p2165_Filter-Quick-Changer-incl--1x-1-25--filter-drawer---low-profile.html) that mounts directly onto my DSLR (previously the filter was taped to the front of the flattener). This roughly halved the distance of the filter to the sensor, to about 55mm.

The star shape significantly improved. When focused I would say the star shape is almost perfect. I could see a slightly non-rounded shape when slightly out of focus each way.

:thanx: to everybody for help :thumbsup: