Jump to content
UltravioletPhotography

[Filter Test] Edmund Hard-coated 340/10 Test #5: Sunflower, Sun


Andrea B.

Recommended Posts

See Also:

[Filter Test 340/10 #1] Gotta work out a problem or two, but here is a 340/10 result anyway

[Filter Test 340/10 #2] Cured the flare but it's ring around the posy here

[Filter Test 340/10 #3] Trying again with the 340/10

[Filter Test 340/10 #4] Edmund Hard-coated 340/10 Rear-mount Experiment

[Filter Test 340/10 #5] here

[Filter Test] 340/10 #6 Progress was made, finally. UPDATE: Progress was then lost.


 

 

 

For my fourth 340/10 experiment today, I picked up a bunch of Sunflowers so I would have something big to shoot at. (Sounds like target practice!!) The sunlight was good this afternoon, so I finally was able to shoot the narrowband Edmund 340/10 outdoors.

 

To review, in the last experiment I rear mounted the 340/10 on the UV-Nikkor with some old Nikon extension parts. With the extension, I've lost infinity focus, but have between 20-30 centimeters of focusing distance from sensor plane to subject. [Filter Test 340/10 #4] Edmund Hard-coated 340/10 Rear-mount Experiment

 

The primary thing to note from today's experiment -- to happily note -- was that I could see through the filter and focus via Live View when the UV-Nikkor was set to its widest at f/4.5. That was a big relief because I have nothing to use as a focusing aid for a narrowband 340. The transmission of the 340/10 is about XXX around its 340 nm peak.

 

The photograph turned out well. (Some flare though.) However, the white balance step gave me some trouble. I'll show you 6 versions. Note that in each version the typical UV-absorbing area at the base of the Sunflower rays and on the central disk is not as dark as usual. This might be because of the characteristics of the UV reflectivity curve of the Sunflower (which, of course, I do not know) or possibly because there is some kind of visible or IR leakage from some area of the camera. I will be checking that out next time I work with the 340/10.

 

 

As Shot: f/8 for 4" @ ISO-400

I use low contrast when shooting UV. To restore contrast, the histogram endpoints were moved in. That is, I adjusted the black and white points. Some local contrast sharpening was added after resizing.

00_helianthus_uvEdmund340x10_sun_20180612wf_10471origBwPtsRes.jpg

 

Raw Composite

Before white balance the photo has these colors. Again, the B&W points were adjusted and local contrast sharpening added after resizing. (A typical raw composite looks very dull and flat without these adjustments.)

01_helianthus_uvEdmund340x10_sun_20180612wf_10471rawCompBwPtsRes.jpg

 

Photo Ninja Conversion

Typically a sunflower can be white balanced on the UV-absorbing areas at the base of the rays. That gave a very strange result here. I'm not sure I consider it a successful WB, but I'm still thinking it over. Some local contrast sharpening was added after resizing.

There is false yellow, blue, green or dark green-cyan and grey in this conversion and the next. That is rather remarkable for such a narrowband filter. Remember, we have all been supposing that there would not be much of a false colour range at all.

02_helianthus_uvEdmund340x10_sun_20180612wf_10471pnPetalWbRes.jpg

 

Capture NX2 Conversion, WB1

Same result from WB on UV absorbing area of petal base. Some local contrast sharpening was added after resizing.

03_helianthus_uvEdmund340x10_sun_20180612wf_10471nx2petalWbRes.jpg

 

Capture NX2 Conversion, WB2

This is an average white balance made over the entire frame. It didn't work out well. (Added: I mean that this version seems to look a bit "flat" to me.) Some local contrast sharpening was added after resizing.

04_helianthus_uvEdmund340x10_sun_20180612wf_10471nx2AveWbRes.jpg

 

Capture NX2 Levels on the Raw Composite

This was a surprise. I really like this version. I don't know what made me think to try a combination of raw composite and levels, but it worked out well.

05_helianthus_uvEdmund340x10_sun_20180612wf_10471rawCompLevsRes.jpg

 

 

Edmund Transmission Chart

I have to write for permission to post the 340/10 chart here because it is copyrighted. But you can go to this link and download a PDF which will show you that the transmission is 95-98% in the filters FWHM range.

LINK: https://www.edmundop...andpass-filter/

 

 

Edmund 340/10 Raw Histogram (Linear) from Raw Digger

This linear plot is a little compressed. There are several spikes up to 16000 not shown here.

helianthus_uvEdmund340x10_sun_20180612wf_10509rawHisto.jpg

 

Edmund 340/10 Raw Histogram (Logarithmic) from Raw Digger

The logarithmic plot better shows what is happening in the brighter areas. Note that the Y-axis scale changes when switching to the Log plot.

00_helianthus_uvEdmund340x10_sun_20180612wf_10471rawHistoLog.jpg

 

 

 

Reference: UV Photo with BaaderU

f/8 for 4" @ ISO-400

Here is the typical broadband UV-signature for the Sunflower.

helianthus_uvBaader_sun_20180612wf_10509pn.jpg

 

Reference: UV Raw Composite with BaaderU

There is more natural contrast here than in the Edmund photo above.

helianthus_uvBaader_sun_20180612wf_10509rawCompNx2.jpg

 

Reference: BaaderU Raw Histogram from Raw Digger

It is interesting that in comparison to the 340/10 histogram above, this one from the BaaderU is more spread out.

helianthus_uvBaader_sun_20180612wf_10509rawHisto.jpg

 

 

Reference: Best exposure times for Edmund 340/10 and BaaderU

This is just an estimate because I was not controlling the variables, but I noticed that both filters shoot about the same in terms of exposure settings. For each filter an exposure between 4-4.5 seconds at f/8 and ISO-400 produced a brightness histogram near in the center. Going to 5-seconds started to blow the Red channels more.

 

 

 

.

Link to comment
In the lower right corner, that is a medium yellow 50/51/52° with a saturation between 27-35%. The tips of the rays have bright yellow 50/51/52° with low saturation between 10-15%. Those samples were made on the TIF. Stuffing the file into a JPG might have altered the sample values.
Link to comment
Interesting, only 2 years or so for you to get this to work. I actually like the WB over full frame shot. Has similar look to Baader venus filter. The purplish is nice, but I know you don't like magenta and its getting there.
Link to comment

I kind of forgot about the 340/10 filter for a while !!

 

What troubles me about the three conversions above is that they seem a bit "flat". One of the major problems I had with the 340/10 when used front-mounted was that it produced so much contrast destroying flare from its mirrored surfaces. Looking at the conversions, I'm still not sure the rear-mount has entirely cured that flare problem. The average WB conversion seemed to look flatter than the other two. But I have been staring at these for too long and my eyes have probably gotten tired. So it is good to hear that one of those conversions looks good to someone!

Link to comment

Seems good for only 335nm to 345nm! The lack of monochrome and yellow colour is actually quite good. As if the software knew it was a sunflower and did its best.

You must remember how restrictive this filter is. I wasn't expecting much more out of it than just green, as seems to be in the "green uv channel".

As per Cadmium spectrum:

http://www.ultravioletphotography.com/content/index.php?app=core&module=attach&section=attach&attach_rel_module=post&attach_id=6991

 

 

Link to comment

Andrea, Nice work.

 

Ulf, Interesting. I looked at that OD graph for a while.

OD3.5 is getting close to the absolute danger zone (OD3), but who knows.

One could test the difference by stacking with Baader U.

View the PDF that Ulf linked to above, scroll down to the bottom graph.

You need to read this graph upside-down from most graphs. Strong transmission is at the bottom, no transmission is at the top, and danger is below OD3 (between OD3 and OD4).

I am not saying the filter leaks or pollutes the 340BP10 band, but it is worth testing, and easy to do..

Link to comment

I agree.

 

Also as the pass band is rather narrow it will pass less light than a filter with a wide passband like Baader U.

The short wavelength and narrow bandwidth of the filter makes the image more sensitive to IR-contamination.

Even the rather narrow dips in the blocking might be enough, especially if the object is IR-bright.

Link to comment

Thanks everyone for the comments! Greatly appreciated. It is very helpful to have other eyes on this. I don't want to get tangled up in Confirmation Bias as in: Because I believe this filter still flares when rear-mounted, then that is what I'm seeing.

 

[[ Off Topic: omg, do go read that list of cognitive biases in the WikidPee!! It is truly amazing that our silly human brains are able to accomplish anything scientific at all. :lol: :lol: :lol: ]]

 

Some Notes:

  • A companion BaaderU raw composite photo was added to the first post. It has much more natural contrast than does the Edmund 340/10 version.

  • The "best" exposure times for each filter was noted. At f/8 on the UV-Nikkor and ISO-400 on the D610, that would be 4-4.5 seconds for each filter. They are both very high transmitting. By "best", I mean that this placed the monochrome brightness histogram as much in the middle as possible. This often can blow out the red channel, but on the D610 it is recoverable.

  • I added raw histograms for both filters. I'm going to repeat those here because they're interesting.

  • Finally, the transmission chart for the 340/10 was linked. It is copyrighted, so I didi not feel like I should add it.


 

The Raw Histograms

 

One interesting observation is that the 340/10 raw histogram has a narrower spread than does the following BaaderU raw histogram. I am thinking that this does indicate that veiling flare from the mirrored surfaces is still a problem even when the filter is raw mounted. So I have a plan to test that further.

 

If anyone is adept at reading raw colours from the histograms, you will note that the narrowband 340/10 produces orange hues and that the broadband BaaderU produces red tones. This is easily confirmed by poking around the raw composite photos with a color sampler.

 

Edmund 340/10 Raw Histogram

helianthus_uvEdmund340x10_sun_20180612wf_10509rawHisto.jpg

 

BaaderU Raw Histogram

helianthus_uvBaader_sun_20180612wf_10509rawHisto.jpg

 


 

 

Plans for Next Experiment with 340/10

  • Shoot Sunflower together with Labsphere Reflective Standards

  • Shoot aperture series to determine whether stopping down past f/8 might help decrease flare -- if flare is indeed present. (How can it not be??)

  • Shoot 34/10 stacks using various Longpass filters to test for visible leakage and/or IR leakage at the "weaker" points on the density chart.

Link to comment

"Finally, the transmission chart for the 340/10 was linked. It is copyrighted, so I did not feel like I should add it."

 

OK, I will remove my copy of it above.

 

A 340BP10 shot with the Baader U added on the front will subtract any 'possible' out of band leaks that may or may not be mixing into the photo.

That would clear up any question about that.

Link to comment
Yes, it does seem strange that Edmund would copyright those transmission charts. You'd think they would like them to be reproduced so that people would become interested in their filters.
Link to comment
  • 1 month later...

Andrea,

I was testing a similar 340bp10 filter I have now at 25mm. These lights that I got cheap from Amazon work well for me:

Sunlite SL20/BLB/3PK 20W Spiral Energy Saving CFL Light Bulb Medium Base (3 Pack).

Using 2 bulbs as close to the subject as possible, but just out of frame using two desk lamps. I got a decent image.

Exposure on my SD14 was ISO 200, f8, 30sec. I was using the 340bp10 filter stacked with Baader venus 2 in front of the lens for exposure. With the lamps directed away (bulb facing subject) I didn't see hot spots or flare.

The exposure on my full spectrum E510 was ISO 1600, f8 50seconds. But I typically get 4 stops better exposure with my Sigma to the Olympus. Your Nikon I think is more sensitive than my Olympus, so these lights may work for you.

Interesting, I had the same exposure setting using sunlight, outside with both cameras. The lens I was using was the Pentax UAT.

I haven't tried mixing flash with these bulbs yet for better exposure times.

The stock EM1 can't see this low. But can see through my 370bp10 filter.

Link to comment

Try stacking the one that you can 'see through' with some longpass, even a common red #25A filter to see if you still see the same through them both.

Also, don't use the sun or UV light, I know you know that, but for others listening in on the conversation...

Link to comment

@Cadmium,

What I am not supposed to stare at the sun? Actually I surprised at how few reports of people staring at the sun came in during the August eclipse.

I also at the same time tested my El 80 nikkor and it was 2 stops slower with 340 band pass and 1 stop slower with 370 filter. If what I was seeing was IR. I would expect no change. The igoriginal 35mm could go to 340, but was hazy. A hood would be needed. My Olympus 35mm macro didn't test well.

But you're correct, throwing on a 720nm long pass would be better to rule out IR contamination. I also want to test the 2mm S8612 filter with 370, as may be better than the baader stacked.

I am going to use the 340, 370, 390 filters I have for some uv tests to see if theere are subject differences. I think one flower did have slight differences, with fine dots in 340. But will have to repeat to be sure. In that test I overexposed the 370 shot slightly.

Link to comment

The red you are seeing may be lower than S8612 will block.

S8612 2mm thick blocks 680nm+ (minimum OD3).

Even at 3.5mm thick it doesn't block the entire red range.

Stacking with the Baader U will remove any leaks above 400nm.

 

post-87-0-15913700-1531705084.jpg

Link to comment

For the 370bp15 filter you may be correct.

Here is an updated spectra for my filter. Clearly not OD4 to 1100nm.

post-188-0-01260700-1531801048.jpg

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...