Jump to content
UltravioletPhotography

Problems with small dichroic filter


Andy Perrin

Recommended Posts

Andy Perrin

As Andrea noted over in the other thread, I was having some major issues with "glow" around the edge of buildings (typically near the top where the sky is) and around the edges. I was using the following setup:

 

Sony NEX-7 body

FotoioX M42-NEX adaptor

Steinheil Munchen EDIXA-Auto-Cassaron 2.8/50mm

49-52mm ring

52-28mm ring

28mm ring with the unthreaded Omega 330WB70 mounted in poster putty inside it.

 

The Omega filter is "black" glass on the rear and dichroic in front.

post-94-0-69010900-1460478198.jpg

 

The rear of the filter looks like this:

post-94-0-35528000-1460478644.jpg

 

The images I'm getting come out of the camera looking like so (this is an especially awful example):

post-94-0-88662900-1460478832.jpg

 

I have just been cropping out the center and working with that, using the panorama function in Photoshop as needed to get more pixels.

Link to comment

There might be light leaks or unintended fluorescence going on here? I'd call this flare although its intensity is pretty low.

 

First, try to find leaks by standing in a dark room with a small yet powerful torch flush into the inner side of the filter. Is there any sign of leaks around the perimeter of the filter?

 

If not, try shining a UV torch on the filter to see if there are areas that light up or indicate something leaking through.

 

My hunch is the putty fastener is the culprit in some way. Try covering the front of the lens with thick black fabric with a hole just big enough to push the very filter through, thus letting fabric covering the putty rim. Does that get rid of the flare?

Link to comment

Totally agree with Bjørn!

 

I think the light is getting in around the dichroic through the uncoated edge. In your first photo I think I might actually see the uncoated edge of your filter. Is that a 25mm filter? If so you could also try cutting an ~20mm hole in some Al foil and mask off the front of the filter so that the edge is covered.

 

PS

Check out Enrico's page on Using special filters and lens shades

Link to comment
Andy Perrin

Thanks, will try those. JCDowdy, I did put aluminum foil around the front of the filter covering the putty (you can see it above, actually). The putty totally surrounds the filter to a depth of several millimeters, so I don't think there is an "uncoated edge" since it's so thick. You may be mistaking the filter's black unthreaded ring for a gap? At the moment, my money is on the fluorescence rather than a light leak, but I will test both.

 

Enrico is saying black silicone instead of the putty, so that may be my solution here, if it is the fluorescence.

Link to comment

Enrico is saying black silicone instead of the putty, so that may be my solution here, if it is the fluorescence.

 

Permatex ULTRA BLACK 0.5 OZ is handy, got mine at AutoZone.

Link to comment
I'm eager to learn if this solves the problem. So be sure to let us know, Andy! (When you get the time, of course.)
Link to comment

Just thought of something. Remember a few days ago the reflection problem I had with the BaaderU on the Vivitar 35/3.5 on the Sony a7R?.

Could the putty be causing a reflection problem like that instead of it being a light leak problem?

 

Here are two of the bad shots again so we don't have to dig out the old post. They were made against a black canvas, so there is no subject. Just reflection rings.

 

BaaderU: pink side out.

a7R_filterTest_baaaderUpinkOut_20160329wf_143458amp2_X.jpg

 

BaaderU: green side out, worse.

a7R_filterTest_baaaderUgreenOut_20160329wf_143514amp6_X.jpg

Link to comment
Andy Perrin
Yeah, could be that too. I already bought the black silicone so I'm just going to replace the putty and see if the problem goes away. By the way, I have preliminarily confirmed that the putty DOES fluoresce under my 365nm flashlight. Whether that is the cause of the difficulty or whether a reflection is causing it remains unknown.
Link to comment
I am assuming that the center of the image is the "good" part, i.e. what is desired, and that the periphery is the "bad" part. I find it hard to believe that fluorescence could project such a sharp-edged and even pattern around the periphery; I would expect something a bit more vague or diffuse were that the case. I would bet that this is a reflection artifact somehow dependent on angle; dichroic filters do produce those sometimes, especially with specular reflection off sensor surfaces..
Link to comment

I experienced a similar look with a 'Tri-band' filter, having one band in the high red, and two more distinct bands in the NIR.

The filter was framed and lot leaking light. I blamed it on the dichroic coating changing the color from center to sides, as sometimes experiences with UV/IR-Cut 'hot-mirror' type filters.

I no longer have the filter, I sent it back.

It will be interesting to find out if the better sealing of your filter will resolve the color shift around the edges.

I was quite excited about the Tri-band filter before I used it.

Here is an example shot (52mm filter, 18-55mm VR @18mm)

post-87-0-22408300-1460520607.jpg

Link to comment
Andy Perrin
Cadmium, putting your experience and Andreas together, I am starting to wonder if changing the sealant WILL fix it. We shall see, I suppose. Your "defective" filter produced kind of a pretty picture, though! Maybe you should have kept it as an art filter.
Link to comment

Cadmium, I am wondering about this image. It would almost appear to be some kind of extreme lateral chromatic aberration because the lens could not focus the different wavelengths to the same plane. This would not be surprising in an "ordinary" lens which has been appropriated for UV or IR or mixed wavelength usage. The thing to do would be to try this triband filter on a lens like the CO60/4.0 which can focus all wavelengths in the same plane. However, you no longer have it. :D May I ask exactly what filter this was/is? I might try one.

 

Of course, I am speculating here because I do not have deep knowledge of how the various aberrations appear in our UV/IR work.

 

Andy, if the new sealant does not improve things, you could maybe try a rear mount for this filter??

Also, is that Sony NEX a full frame? This filter might work better over a smaller sensor??

Link to comment
Andy Perrin

Andrea, i

Andy, if the new sealant does not improve things, you could maybe try a rear mount for this filter??

Also, is that Sony NEX a full frame? This filter might work better over a smaller sensor??

The Sony NEX-7 is APS-C. So, middle of the road I guess? Crop factor of 1.53. And it's funny you mentioned rear mounting the filter because I just tried it and it works! I am guessing that the rays that make it to the back of the lens are already pretty paraxial[*] so if this is being caused by the dichroic behaving oddly at large angles, that would explain the issue.

 

[*] showing off fancy optics vocab :-)

 

 

Link to comment

Good fix, I like rear mounting when possible, especially with small thicker filters. However that makes swapping filters difficult unless a filter drawer is used.

 

It might be instructive to test the effect of a deep/conical lens hood on the front mount orientation.

Link to comment

Cadmium, I am wondering about this image. It would almost appear to be some kind of extreme lateral chromatic aberration because the lens could not focus the different wavelengths to the same plane. This would not be surprising in an "ordinary" lens which has been appropriated for UV or IR or mixed wavelength usage. The thing to do would be to try this triband filter on a lens like the CO60/4.0 which can focus all wavelengths in the same plane. However, you no longer have it. :D May I ask exactly what filter this was/is? I might try one.

 

Of course, I am speculating here because I do not have deep knowledge of how the various aberrations appear in our UV/IR work.

 

I don't have a CO60/4.

Search for "dual band filter" and "triple band filter" on eBay, you will find a few. I am not sure if they have the exact same two I tried.

At least one of the "dual band filters" is actually a "tipple band filter".

They should work fine with 60mm, or pretty much any lens of that focal length, this was shot at 18mm, so anything above a certain focal length will avoid the altered edge color. That is how it worked for me, wide angle had different colored edges, longer focal length showed just the center colors. I had two, one was 50mm the other 52mm. I don't see the exact same items numbers currently listed.

Link to comment
Andy Perrin

Ok, here's with the back-mounted filter.

 

House from yesterday

(330WB70 back-mounted, Novoflex Noflexar 35/3.5, F/3.5, 1/60", ISO6400)

post-94-0-97832100-1460591896.jpg

 

Dandelion

(330WB70 back-mounted, Novoflex Noflexar 35/3.5, F/3.5, 1/100", ISO12800)

post-94-0-51540300-1460591671.jpg

Link to comment

Looking much better. I think you mostly cured the problem.

 

Try also adding a lens hood as JD suggested??

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...