Universal lens combination with color match for travel, landscape and street

[Kai]

During the summer holidays I had a modern Canon EF 1: 1.8 50 mm and an old 1: 3.5 28mm lens (M42, "Accura") with me. My camera was my EOS 6 D-FS (for non Canon users: it is full format). When processing the images, it became apparent that the different color behavior (due to the different color depth and the different filters that I used) was very annoying.

 

So I was looking for two full-frame lenses that I can use at the same time when traveling. The focal lengths have proven for me for landscape and street. For macros, I would rather stick with my enlarger lens.

 

Enclosed the solar spectrum, taken with a 1:1.8-50 mm II and the newer 1:1.8-50 mm STM and an old 1: 2.8 28 mm (non USM).
Canon EOS 6 D-FS, sunlight (grayish blue, cloudless sky), slit, DVD grating, Baader-U, white balance on PTFE.

 

 

For comparison, an image through an enlarger lens:

 

 

Results
The two AF lenses transmit light from about 395 nm down to 365 nm.
The color change takes place at around 378-380 nm.

The color rendering harmonizes well with each other. 
Both lenses can be used with AF (works slowly but works!).

Both lenses are light, inexpensive and can also be used for other purposes.
For good imaging performance, however, they must be stopped down to at least f / 4, better to 5.6, 8 or 11. 

Of course, with such a low UV depth, yellow tones are only rendered very pale in sunlight. 

 

Example pictures
(The green spots on the left in the picture are IR-artifacts and sometimes arise when taking photos from the live view. )

 

Germany, August, 5 pm, Baader-U, EOS 6D-FS, EF 50 mm, f 4.0, 1/30 s, 3200 ASA

 

 

August, 5 pm, Baader-U, EOS 6D-FS, EF 28 mm, f 4.0, 1/15 s, 3200 ASA

 

 

Today I would not use a filter with NIR interference suppression like the Baader-U for 28 mm, as the strong angle dependence of the filtering effect is clearly annoying. An additional QB21 helps...

 

[dabateman]

Yes that's also what I see. My Canon 50mm f1.8 ii had a uv cutoff at 363nm before I drilled through the front 3 elements. The back 3 and front 3 have the same transmission.  You can pop out the back 3 elements easily to get a 100mm lens,  but you will need to place it on extension tubes,  as it will not focus within the front barrel range.

 

My STM 50mm f1.8 wasn't as good, as the ii. But similar as you have shown.  I think my cut off was 365nm.

 

The best for Canon so far is the 40mm STM f2.8 lens.  Its cutoff is about 350nm, and focuses very well in UV. One of my favorite autofocus lenses.  It also sees far into IR, without coating blockage.  So a good all round lens. You can also reverse it with autofocus macro reverse mount adapter and get 1:1 Macro. 

 

[Andrea B.]

Kai, thank you for the Canon lenses recommendations for UV false color matching.

 

((Side Note: Be sure to toss in the "Canon" brand when naming lenses in photo captions because not everyone would recognize that "EOS" indicates Canon. ))

 

Over time it seems that most of the UV false color images I've made look similar after white balancing against a known "white" standard for UV. But if you are seeing some false color differences between different lenses when photographing the same scene, then it would be interesting to see an example if you have one handy??

[Kai]

Hello Andrea,

First (and that's what this post should be about) I looked for lenses with different angles of view that have the same transmission curves. The idea was that I don't have to redo the white balance after every lens change, but can freely change the lenses - as long as the lighting situation does not change. This makes taking photos while traveling a lot easier. So this is about color matching together with white balance.
As far as pure (false) color reproduction is concerned, it is probably the case that the color differentiation is better the farer a lens reaches into the UV and the larger the recorded wavelength range is. This is probably a consequence of the absorption spectra of the dyes in the Bayer mask. With my cameras, the differentiation is only lost at below 340 nm. Then every wavelength appears green.
In principle, all Bayer dyes seem to lead to the same colors (green, yellow, colorless, violet), the exact position (hue-value and wavelength in case of monochromatic light) seems to me to depend on the individual transmission curve of the lens in combination with the filter used (and the light source).
What do you think?
P.S. I can look for sample pictures when I am back home in some days.