My first UV photograph

[Andrea B.]

Jim: I was going to download rawdigger to try but McAfee thinks its a virus and advises against it - so I didn't. Anyone else found this to be an issue?

 

Well, not that I've ever heard. But I'm on a Mac so don't run things like McAfee.

Anyway, I'd write to the Raw Digger guys. They are very responsive.

 

QUESTION 1: I assume this image is dominated by near IR with significant visible if I have a full spectrum conversion - These wavelengths will dominate UV completely ?

Not sure we have a precise way to determine whether the Vis+IR dominates "completely". But there is likely very little UV being recorded by an unfiltered lens over an unfiltered sensor. This is primarily due to the composition of sunlight. I'm going to go with a relative percentage ratio of 3/43/54% for UV/Vis/IR light in sunlight. This of course varies by altitude, latitude, time of year and other factors. The UV may go to about 4%. Please view the solar composition chart: http://www.ultraviol...ation-spectrum/

 

To convince yourself of all this it is a very good exercise to shoot a subject with a known UV-signature under various filters. At this time of year a Sunflower from the grocery store or the florist works well. (Not sure where Sunflowers are found in January in the UK. Here in the Colonies we would look in the grocery store.) A Sunflower has a very dark, UV-absorbing central disc and a very dark, UV-absorbing band on the petals around the disc. The outer (distal) portion of the petals is very light and UV-reflecting. Adding a Sunflower to your test scenes will soon tell you how much UV you are capturing in an unfiltered shooting scenario. If the central bullseye is "washed out" or almost non-existant, then you are not recording much UV.

[Andrea B.]

QUESTION 2: It seemed to me that the Soligor lens (top image) is a little warmer red and therefore maybe a little better for UV? (If anyone is interested the nef file is here

 

As mentioned earlier, it is difficult to say because filtration also plays a role in what is recorded. I will look at your Soligor image in Raw Digger and post here.

 

Please go view this topic: Three Sunflowers - Three Filters [baaderU, KolariU, StraightEdgeU]

Scroll down to the raw composites.

Note the basic color difference between the orange BaaderU/KolariU fotos and the magenta/pink StraightEdgeU foto. Fotos were made with Nikon D610-mod and UV-Nikkor 105/4.5 which has a bandwidth of 200-1100 nm.

 

The StraightEdgeU is a UV-pass filter designated 379fwhm52 which has a very steep right shoulder which passes no violet. The StraightEdgeU's raw composite shows us that when we are recording within 20nm of that approximate 380nm peak, then we are going to see mostly raw magenta/pink colors. We always get lots of UV recorded in the red channel. Magenta tells us that there is also a lot being recorded in the blue channel, moreso than in the green channel.

 

The BaaderU UV-pass filter is designated 350fwhm60 and the KolariU as 365fwhm40. Their raw composites shows that when we record in a 20-30nm band around these lower peaks, we are getting mostly raw orange colors. I.e., the green channel is picking up more UV.

 

That little orange/magenta analysis is predicated on the way our current set of sensors and Bayer filters respond to UV light. I certainly cannot say that this orange versus magenta thing will hold in the future.

 

---

 

 

Yes, I think you are recording a little bit deeper in the UV with this Soligor because the blue & green histos are more nearly alike. But do work on getting these exposures up by at least a stop. Too much noise here.

[Andrea B.]

QUESTION 3: what is this - is it dust on the sensor? If so, I guess its a problem if its behind the new window?

 

No, not sensor dust. That shows up blurry and dark. The bright dots might be noise or might be hot pixels. I'm not really sure what they are. Is that dark area on the tree painted?? Why is it so dark? If that is paint, then maybe we are seeing something about the paint or whatever.

 

I'll look at the raw NEF if you give me a link for it. That might be enlightening about the dots.

[Andrea B.]

QUESTION 4: What is making things blue in the image - seems particularly man made things. I assume this is because they are absorbing shorter wavelengths of UV and reflecting just the longer wavelengths? is that right?

 

Some metals are false blue/violet in UV. Some kinds of paint also -- with titanium I think.

 

*********

 

I am really enjoying the artistry and composition of the fallen birch(?) log photograph. The tones and false colours hang together so nicely. So here is a question for you. Would this little composition also be as interesting in Visible or IR light? Sometimes we get a nice look in UV which may not translate to other wavebands. Usually, it seems, the opposite case holds (!) - what's pretty in Visible light looks somewhat scrungy* in UV. I have long thought that UV photography might be an acquired taste rather than one which happens naturally.

 

 

 

^{*scrungy -- this is not a formal art or photography term. :rolleyes:}

[Andrea B.]

Jim: Let me know if I am not following the correct forum etiquette regarding number of images or use of threads etc

 

As for long threads, if someone begins to complain about the load time from the server then it might be nice to start a new topic after 3 or 4 pages. We do not currently have the faster server option so topics a few pages long might slow-load. I don't see this happening currently with my set-up. :)

 

And I don't mind threads which wander from the topic as long as they are not in the botanical section which is meant to be a kind of resource.

[Andrea B.]

Ulf: Nice mini-series! Yes, it's a good illustration of the diffs. The blackberries are so wonderfully shiny in UV. They look like they are made of metal, don't they?

[Jim Lloyd]

Hi Andrea

 

Thanks for all your replies - very helpful as always :) ! . I am just about to go out, so will respond more tomorrow.

 

Re the hot dots: here is a link to the nef file. Its moss on the tree trunk which is bright green in visible light I think the same feature is on many of the other images, but only really noticeable in dark areas - so maybe you are right some sort of noise? Its a relief its not anything more serious.

 

The question re artistry is very interesting and I will reply more tomorrow - I could say quite a lot (and maybe bore people!. )

 

I have photographed these woods a lot see here and here (and listen here ) - I am interested when art becomes more than just presenting an image, but becomes about an intense exploration of an area - and then one can question whether its art ...

 

I look out for a sunflower - the only flowers growing around here at the moment are snowdrops and some gorse which strangely seems to keep flowering in any season

[Andrea B.]

Snowdrops and gorse!! Yes, fotos, please!!

 

**********

 

Jim, I find the odd points of bright to be a bit troubling. That is a strange kind of noise. Granted there was very much noise in the foto I examined above, but still.

 

Does anyone still reading along have any suggestions about these pinpoints of noise?

Scrolll down in post #50 above and look for the dark crop with the bright specks: http://www.ultravioletphotography.com/content/index.php/topic/2543-my-first-uv-photograph/page__view__findpost__p__19070

[Andrea B.]

Went out for a quick grab shot using D610 + UV-Nikkor 105/3.5 + UG1 (2mm) +BG40 (2mm).

I wanted to see what the raw composite and raw histograms would look like with a wideband lens.

 

From my results below you can see that this filter combo on a wideband lens is recording a magenta/pink raw look with Blue and Green channels about equal. This implies that the filter combo is perhaps passing some violet or recording more to the right of its proposed peak of 365nm. The charts above would suggest that a violet leak is the reason we are getting the raw magenta/pink.

 

Remember that in some cameras actual spectral violet can only be recorded in the blue channel and in other cameras spectral violet is recorded (like eyesight) in both the blue and red channels. Lacking authoritative spectral charts for any of our modified cameras, we should probably simply leave spectral violet's channels speculative for now. And finally, there really is no hard spectral line where any of our well-known UV-signatures suddenly disappear -- they tend rather to fade out gradually. A bit of violet does no harm.

 

Corrections and suggestions always welcomed!

 

f/4.5 for 1/2" @ ISO-200

 

Raw Composite from Raw Digger

 

Raw Histograms from Raw Digger

^{I hit the right red wall ay 16000, but also underexposed a bit. Oh well. That winter light is indeed tough.}

 

Rendered Scene

^{boring! "-) Kind of a dull day, flat light.}

[OlDoinyo]

 

 

Remember that in some cameras actual spectral violet can only be recorded in the blue channel and in other cameras spectral violet is recorded (like eyesight) in both the blue and red channels.

 

 

 

In every camera I have encountered, the "red" channel was most sensitive to UV, and was presumably engineered that way partly to mimic the bimodal sensitivity of the human "red" visual channel (after the UV-proper wavelengths are blocked.) Have you encountered any camera that behaved otherwise?

[Andrea B.]

The typical CCD and CMOS sensored cameras I've used so far are like that in the raw files, for sure! Red predominates. Almost overwhelms. This is when using broadband lenses and UV-pass filters.

 

****

When the human eye receives a spectral violet signal all of the R, G and B pigments are stimulated but the brain ignores the green stimulation. However, the human eye does not see blue as well as red and green and sees spectral violet even more poorly. And sadly our aging eyes with their yellowing lenses block even more blue and violet.

 

The camera cannot ignore any green signals as it tries to mimic our color vision. For spectral violet, the camera records a signal somewhat like (R=n, G1=n, G2=n, B=2n) for spectral violet ((that is, much more blue than red and green)) which is decoded as a kind of blue (a "greyed" blue). The Bayer pigments vary a bit across the camera brands. So this signal might be decoded with more or less red or green. Then the role of gamut comes into play. Cannot fit spectral violet into monitor gamuts.

 

{{I am very aware that this explanation is not precise and has no references. Please investigate this as it is all quite interesting to read about! There is a lot out there about why digital cameras have problems with both non-spectral purple and spectral violet.}}

[ulf]

Ulf: Nice mini-series! Yes, it's a good illustration of the diffs. The blackberries are so wonderfully shiny in UV. They look like they are made of metal, don't they?

 

Thank you Andrea

 

Yes they do look like a very hard high-tech metal.

 

I have four other filtration variants of the same motive, not fully processed yet.

Visual (BG38, 2mm); [s8612, 2mm + UG5, 1.5mm]; [s8612, 2mm + BG3, 2mm] and [s8612, 2mm + BG25, 2mm]

 

When ready to publish Is it OK to start a new thread and reuse the ones above?

I think the full set has a separate value, but the three above are meaningful in this thread.

[Cadmium]

Ulf, That's a really nice visual demonstration.

Thanks Steve

I guess what I am using should be more like image 1 or 2 that image 3.

 

Jim, No, your stack is not like Ulf's image #1. He is using UG1 1mm thick, which is thinner than yours, thus sliding the cut-off point further over into the blue.

Your UG1 2mm stack would look more similar to how his Baader U image is, with less blue than his image #1.

Your UG1 stack is really not designed to have much blue in it, Ulf's image #1 stack is designed to have a touch of blue, and what you see is the result.

Guess I will make a graph comparing these...

[Cadmium]

Jim, For UG1 2mm only shots, what I call 'dual band IR', try white balancing a small area on the foliage, see the arrow in my CNX2 demo:

http://www.ultraviol...dpost__p__18887

This will keep your foliage as white as possible, and change the color of the sky.

You are using UG1 2mm, and I was using U-360 2mm, but these are fairly close, so yours would look closer to what I show in my demo.

Try different small spots/areas. I usually select the most colored foliage, this will get the foliage more white.

[Cadmium]

Here is the graph.

It is a little hard to read given the six plots, but let me point out a couple things.

The Black (UG1 2mm + BG40 2mm) and Purple (U-360 2mm + S8612 2mm) plots share almost identical UV transmission, and they are hard to read from each other on the grap as a result,

however, the Purple (U-360) plot shows it has slightly less blue than the Black (UG1) plot.

The real difference between those two stacks is the Red/IR suppression.

Also, notice the greatly reduced UV transmission with the Green (UG1 2mm + S8612 2mm) plot, now compare that to the Purple (U-360 2mm + S8612 2mm) plot,

and you may understand why I recommend U-360 instead of UG1. If you want an even stronger UV stack, then use U-360 2mm + S8612 1.5mm (the Light Blue plot).

So, my pick of the bunch would be the Purple and Light Blue plots, both U-360 2mm, using S8612 1.5mm or 2mm. Both of these stacks keep the blue cut good, and the peak UV close to 360nm,

both have ample or more Red/IR suppression. Want more blue, then change the U-360 2mm to 1mm, use the same thickness of S8612.

NOTE: Calculations are only as good as the raw data, and Hoya doesn't provide quite adequate data for U-360, data drops off above 410nm, so any U-360 1mm plots will not show the extent of blue that is actually present.

See: http://www.hoyaoptics.com/pdf/U360.pdf

[Andrea B.]

Ulf: When ready to publish Is it OK to start a new thread and reuse the ones above?

 

Yes, that is quite OK. :)

[Andrea B.]

Something to keep in mind with the filter transmission charts is that your lens will be cutting some of the transmission on the left and also shifting the transmission peak to the right unless the lens is broadband enough to include 300 nm. Not many lenses can "reach" that far.

[Cadmium]

However, most UV friendly UV lenses, similar to a Kuribayashi 35mm (for example), etc., are deep and inclusive enough NOT to change the peak of a 360nm filter.

Try comparing any of these filters between a Kuri and a UV-Nikkor, and show me the real difference.

Of course, lenses that are not exactly UV friendly, such as a Nikon 18mm f/4, or a lot of old Nikon lenses, even an E lens, would definitely truncate the UV filter curve some, shifting the UV peak transmission upward, and also depressing the amplitude of that peak as well. Such lenses are still usable of course, like the 18mm f/4, which is one of the best 18mm UV lenses.

So your lens is very important.

Let me see if I can find an illustration of all that...

 

OK, so here is an example.

So there are many lenses, other than a UV-Nikkor, and similar to a Kuribayashi 35mm f/3.5, that would not truncate the UV transmission of the typical UV-A range filter.

[Jim Lloyd]

Thanks Steve

 

I wonder what difference changing from 2 to 1 mm makes in absolute terms? I guess doubles the absolute transmission. I noticed yesterday that I only just had enough transmission to be able to use live view to compose shots- certainly not enough to properly judge focus. So I was thinking maybe a thinner filter might be good from that aspect. I think for artistic reasons a little visible leak might be ok.

 

I guess the difficulty just looking st the filter chart is that you really need to multiply by the ambient spectrum and sensor and lens response to know how the image is affected

 

Sun is out today !

[Jim Lloyd]

I have just realised I am being a bit dim !

 

The lens I was using had pre set aperture so I should open it up to compose and then stop down to shoot. I am relying on focus scale for focus and it seems to work. I want to avoid removing filters in the field

 

[Cadmium]

Jim, The reason I am here in the topic is to help you white balance.

You seem to have figure that out.

Beyond that, this topic has taken many obtuse turns.

Your question above, if you change 2mm UG1 or U-360 to 1mm it will make the result include some more visual blue.

Live view is often hard to use. Best to use a torch (that Bob Friedman still likes to call a flashlight, which is maybe more of a USA term, but so hard to say and type compared to torch).

I use a torch sometimes to illuminate the focus point.

Also Bob taught me the other day that one can set up a button on the back of the Nikon to zoom in live view to max for focus.

I have not set this up yet, but it sounds like a very useful tool.

With a lens like the Kuribayashi 35mm f/3.5 you can remove the UV filter and focus with visual, because the focal shift between visual and UV is nominal with the Kuri.

 

No, a thinner filter will only make it transmit more blue.

You are fine, just stay with what you have. It is working for you great. Unless you would like to swap for other filters...

Like I said, the only thing you need to keep an eye out for with the stack you are using is the red/IR leak.

 

You are doing great.

[Andrea B.]

JIm, I agree with Steve that you are doing great! It is simply practice that is needed now. :) We have probably overwhelmed you with all this advice and data. But read through it a step at a time.....etc etc. ;)

 

I tried your filter stack yesterday (look in Post#59 ) and it is quite OK. Later on you can try another type. None of us ever have just one UV-pass filter or filter stack !!!!

 

Focusing in UV has its problems, sure, but it can be done. You will work out the best procedure for yourself soon. Remember that "focus bracketing" is a tried-and-true methodology for UV scenarios. The very worst case is that you focus wide open in Visible light, add the UV filter, stop down and then shoot at that focus and at several notches before and after that focus.

 

As for not wanting to change, remove, replace filters in the field while shooting.......there is one old solution which I'll look for and make a link to for you. Otherwise, you just have to be careful. I usually hold one hand under the filter while removing it with the other. (But I have dropped filters into mud, dust, grass, pavements, onto rocks and so forth -- along with cameras and lenses. It happens.) You could also use one of those lens cap solutions. There is a sticky button which sticks onto the side of the filter attached to a loop which goes around the lens. So if you drop the filter it dangles from the lens. That would work.

 

The following links were chosen at random to illustrate the product. Shop for the best price.

Lens Cap Leash - https://www.amazon.c...x/dp/B003U7URH6

Old Nikon Filter Holder with Swing-away Door - https://www.amazon.com/Nikon-Gelatin-Filter-requires-adapter/dp/B0000BZM4R

[Jim Lloyd]

Thanks again Andrea and Steve, although the topic has wandered a bit, its been fantastic for a newby to learn! - as you say a lot to take in - I have already been back over the thread several times to look again at advice.

 

I think I may try the 1 mm UG1 to add in a bit of blue later, but stick to what I have for now. My experience with photography generally is that it is good to let one thing bed in before moving to the next.

 

I may start another thread that picks up some issues in this one regarding the art side of UV photography. I am just putting together something from today's shoot to show.

[Cadmium]

*** BEEP BEEP!!! WARNING WARNING!!! ***

Do NOT use a UG1 1mm + BG40 2mm filter stack.

It doesn't have enough Red/IR suppression.

You will need S8612 2mm (or BG40 2.5+ mm) to preform such acrobatics without a leak.

UG1 1mm + BG40 2mm leaks Red/IR.

 

[Jim Lloyd]

Thanks Steve - very helpful information.