[Filter Test] 340AF15 Strange Results

[Andrea B.]

I made some experiments with Hoya U340 stacks and a narrowband 340AF15 filter yesterday. I wanted to post the series of photos and see what everyone thinks. My purpose was to explore false colours under 340nm using the Nikon D600 and UV-Nikkor. I will just show the 340AF15 results in this thread because I think there will be some questions and comments. I certainly have some! :D

 

Equipment: Nikon D600-broadband + Nikon 105/4.5 UV-Nikkor + Sunlight

Visible Filter: Baader UV/IR-Cut

 

UV-Pass Filter: Omega 340AF15

Specs for this filter can be found here: http://www.omegafilt...0af15-1840.html

The bandpass is from 335-345nm. AF stands for 'alpha filter' in Omega's designations. The filter is 15mm in diameter. So it is useable on the front of the UV-Nikkor with some care.

I have one of the Ebay versions and not the $200 stock version. My understanding is that surface quality differs between the Ebay low-priced 340AF15 and the stock.

 

Added Later: I put the 340AF15 chart in post #4 below.

 

Visible [f/11 for 1/400" @ ISO-100]

Reference shot of the white and purple crocus.

 

Ultraviolet [f/11 for 4" @ ISO-400]

In Photo Ninja, the file was demosaiced, and then the exposure was adjusted as needed. No white balance adjustments were made. No D600 colour profiling was applied. The jpg was saved with sRGB.

This represents Photo Ninja's idea of what was actually recorded - maybe - I'm no longer sure about that. I think PN might be applying some kind of default 'daylight' profile?

 

Raw Digger Screenshot

As a sanity check on preceding conversion, here is a screenshot from Raw Digger showing the raw composite of this file. It differs from the preceding photo. (Note: Currently my RD is only working in trial mode.)

 

**********

 

Here is the 340AF15 experiment repeated with the Color Checker Passport and Labsphere Reflectance Standards.

 

Visible [f/4.5 for 1/400" @ ISO-100]

Reference shot of CCPassport and Standards.

 

Ultraviolet [f/4.5 for 2.5" @ ISO-400]

In Photo Ninja, the file was demosaiced, and then the exposure was adjusted as needed. No white balance adjustments were made. No D600 colour profiling was applied. The jpg was saved with sRGB.

 

Raw Digger Screenshot

As a sanity check on preceding conversion, here is a screenshot from Raw Digger showing the raw composite of this file. Raw Digger is complaining (see top right) that Blue is underexposed. Well, yes, because not much was recorded in the blue channel at all. (Note: Currently my RD is only working in trial mode.)

 

****************************

 

So what to conclude at this point? It appears that Photo Ninja might be applying some kind of white balance or profile even though I have not set anything in the Color Correction tool and have left that box unchecked.

 

But it all gets worse when I try to white balance these photos given that there is little to no blue recorded. Really you cannot white balance such a 2-channel photo if you think about it for a minute. Here's the best I could do.

 

White balance attempt in Photo Ninja using white standard.

This creates a blue tint in the grey standards. If wb is performed on one of the grey standards, then that leaves a yellow tint in the white standards. I gave up.

 

White balance attempt in NX2.

NX2 cannot handle a white balance at all !!!

[Andrea B.]

I think it is imperative at this point that I go pay for my Raw Digger upgrade and get it going again because we really, really do need to see the histogram of the raw data for this particular combination of camera/lens/filter.

[nfoto]

How can you expect 'white' with only 2 colours? Whether you have paid for the software license or no, this is a moot point.

[Andrea B.]

That's exactly what I said !! You can't. Still, I really do want to see the actual raw data.

Photo Ninja certainly gave it the old college try on the white balance thing and did manage to get the white standard almost white.

 

I'm not getting any green.

 

John said:

Well, you may not be getting any green at 340nm because the filter is not centered at 340.

It is really a 334nm ± 8nm, cutting off the long half of what a 340nm filter would normally pass.

Seems to be oddly named more like a Raman edge filter.

 

Here is an enlargement I (Andrea B.) just made from the Omega link above.

The lines represent 0-70% transmission in increments of 10. The trans curve just breaks the 60% line.

So why do they label this a 340AF15 anyway ??

[nfoto]

PN made the target "almost white" by adding the third colour that effectively isn't present. Meaning the target will be (almost) white and everything else overlaid by blue.

 

I got green hues with my BP340 filter. On a D200 or D40 (not D40x) if memory serves. Lens was UV-Nikkor and illumination was studio flashes with uncoated Xenon tubes.

[Andrea B.]

Which BP340 filter is that, Bjørn? (Thx.)

 

I have some adventures to post with my Hoya U340 later.

[nfoto]

Omega, I think. It has been lying around for years so not entirely sure of its origin.

[Andrea B.]

Here's the histo data for the CC/Standards shot. It certainly illustrates the difficulties of shooting UV. The 4 peaks on the R and G/G charts are, I think, the 4 brightest standards. You can see how close I came in the Red channel to blowing out that white standard. There are a few pixels in the white standard which blew, but not enough of them to prevent the WB attempt.

 

[Andrea B.]

And, here is the raw data for the 340AF15 Crocus in Post #1.

It was easier to set exposure on this one, of course.

 

[Andrea B.]

Perhaps the Raw Composite that Raw Digger creates as shown in Post #1 are RGGB and the Photo Ninja demosaic is RGB. That would definitely explain the colour differences between the two interpretations. I'll investigate this and add a comment above to clarify things if this is true.

 

Later: NO. It looks like the Raw Digger Raw Composite is RGB.

[nfoto]

I'm getting this program and will try to make good use of it. I foresee possibilities outside the realm of UV photography. However it won't require much imagination to see the danger of falling into a UV geek techno trap, so promise to be careful .

[Andrea B.]

RESCUE ME !!!!!

I fell in for sure !!!!!

 

But srsly, Raw Digger is useful to see what is "really there". I've also discovered that for many UV filters RD can produce a lovely auto white balance. Then you can export the result as a 16-bit TIFF for further edits if desired. So there's one additional benefit aside from the techno-geek-ness.