Wide angle decent UV transmission lenses?

[DKoch]

I have the Industar50-2 but i need to have shorter focal lengths... I'm looking at the Soligor 35mm f 3.5; are all lenses of this description good for UV or do they have to come from a certain period?

 Also, I need something in the 24mm ( or wider) range; anyone seen anything?

[ulf]

Soligor's lenses was produced by several different manufacturers. You cannot assume they all have decent UV-transmission!

To be sure you must get the exact type that is verified to be decent. That is very important and true for all lenses.

 

At the period similar lenses was branded for many distributors and just as Soligor is a brand name there are others. If you search here you can find topics about that.

The most hyped brand for 35/3.5mm lenses is the Kuribayashi brand. They were also actually producing the lenses, also for other brands. The Soligor KA-type was made by them.

 

However UV reach is sometimes not all that counts.

As the camera sensors lose sensitivity quite fast, into UV, a good transmission above 350nm is normally most important, if you are not trying to reach UV-B with very special filters and light sources.
A lens with a bit less deep UV reach sometimes can have better image quality.

IMHO the Asahi 35mm f/3.5 Super-Takumar is a better overall performer with a better optical design.

 

Have you discovered our technical section where many lenses have proper transmission measurements:
https://www.ultravioletphotography.com/content/index.php?/forum/645-uv-lens-technical-data/

 

Lenses wider than 35mm that cover a full frame sensor is normally too complex optically for a deeper UV-reach.

The 24x36 mm film format is the typical format where most lenses acceptable for UV are found.

 

35mm seams to be the breaking point, then they get progressively worse for UV, due to need for many optical elements, thick glass and more exotic glass materials.

There are a few 19mm and 21mm lenses that are marginally OK for UV. 

If you want to go really wide the 5.6mm circular Fisheye lens from Sunex is an alternative, but it covers APS-C only

[Andrea B.]

DKoch: please look in section five of the Lens Sticky and look through Ulf's work in  UV Lens Technical Data. Everything we know about UV-capable lenses is in those two places! 

[Andrew Dayer]

By coincidence, yesterday I finally got out some old lenses to do some more or less systematic comparisons.

 

As the subject matter is similar, I offer some results here rather than start a new topic.

 

My goal was to confirm which lenses are of mine are no use for UV. This was intended just for me as a look-and-see exercise so I wasn't worried that the composition or alignment of different focal lengths wasn't great and, with hindsight, I might have started with a slower shutter speed (but being lazy I wanted the same settings for all images).

The image sets consisted each of a visible reference, UV torch illuminated and full spectrum flash (modified Vivitar 285HV). All images made on a full spectrum Nikon D3200 with viewfinder covered at ISO 800 and at lens maximum aperture. Baader U filter used for all UV images (Hoya UV/IR block for reference visible images (not shown here)). Light source was Alonefire X901UV 365nm LED torch with stock filter.

Subject is a pot of orchids (which had previously photographed OK in UV and are about all I had to hand). Photographed inside with ambient light through glass windows; later on with halogen spot lights; both assumed insignificant for UV (confirmed by accidental image with no UV light and Baader U). PTFE target used for white balance in Lightroom (LR) using custom profile (DNG Profile Editor; temp and tint both preset to -100). Exposure adjusted in LR to approximate f3.5. No other adjustments except cropping.

The flash images are not shown here; some are rather burned out, some have the UV torch still on (may be no difference but...) and I worry that the results might be IR more contaminated.

 

Old 35mm f3.5 lenses. All M42 thread. Exposure not adjusted (EV~6.5)
 

Optomax 35mm f3.5 s/n 36325. 46mm filter thread; body style as per https://www.savazzi.net/photography/35optomax.html

 

 

Soligor 35mm f3.5 s/n 45055. 46mm filter thread; dirty lens; body style similar to Kyoei Type 2 (https://www.savazzi.net/photography/35kyoei.html)

 

Prinz Galaxy 35mm f3.5 s/n 59583. 49mm filter thread; body style as Photax Type 1 https://www.savazzi.net/photography/35photax.html. Faulty; not tested.

 

[will edit and include if possible]

Unitor 35mm f3.5  s/n 40028. 49mm filter thread; body style similar to Photax Type 1 (with cosmetric changes) https://www.savazzi.net/photography/35photax.html




Helios 44 58mm f2. All M42 thread. Exposures adusted -1.65 in LR to EV~6.5

Helios 44 58mm f2 KMZ factory s/n 0428484. 49mm filter thread

 

 

Helios 44-2 58mm f2 KMZ factory s/n 7411981. 49mm filter thread
 

 

 

Helios 44M-4 Julpiter factory s/n 87376724 52mm filter thread.



 

Some 'control' lenses. Exposure adjustments per lens

Nikon 105mm f2.8 D Macro (actual max aperature at this focus distance f3); adjusted -0.3 stops in LR (EV ~6.6)


 

 

UV-Nikkor 105mm f4.5. Adjusted +0.65 stops in LR (EV ~6.5)




Nikon 18-55mm f3-5.6 VR DX kit lens @ 55mm f5.6. Adjusted +1.33 stops in LR (EV ~6.6)




Comments:

• Overall, I'm not happy with these. Not much of a clear difference in UV performance - does that mean IR leakage? Where from - Baader U's usually OK and the LED torch is not IR rich. Maybe I'm just not deep enough in UV to see important differences?
   
• I was surprised not too see more difference between the 35mm f3.5's. The Unitor 'shouldn't' pass much UV but isn't much different to the other versions.
• The 35mm f3.5's were not easy to focus in live view and didn't ever seem sharp.
   
• I was really surprised to see any UV signal with the Helios 58mm's. At f2, there is of course more signal than the 'stopped' down comparsion images show.
• I've had these for a while but never tried them for anything. But really liked them - they do have a lovely image separation at max aperature. Pretty sharp and easier with live view. I might be using these a lot even if not for UV.
   
• The difference between the 105mm's is also not great. Sure, the 105mm UV is parfocal and a will work far deeper into UV than 365nm but at this wavelength, its hard to justify a 20x price differential...
   
• ... and the 18-55mm is usable with limitations.

Just for reference, here are three images with flash (full power from <1 metre). Same adjustment basis as before:

105mm UV (rather burned out) EV ~6.6




Optomax 35mm f3.5 EV 6.5




18-55mm @ 55mm f5.6 EV ~6.6



If its sunny tomorrow, I might try for a UV landscape with the 18-55mm. Just to see...

Comments and mild rebukes for weak methodology welcome.

 

[Andrea B.]

I will make those 2 go away. One of them was doubly uploaded. Not sure about the other.

(Haven't read your comment just yet. But will do.)

 

So, all the images you uploaded were named "image.png". Thus I cannot figure out which uploaded file is the unwanted last one!

 

later -- Found it.

 

Remove an uploaded file from the upload array by clicking on the X.

[Andrea B.]

The first 9 images were made in ambient window light? That may not be enough UV light to bring out differences between "good" UV lenses and not-so-good ones if all are getting only light above 380 nm?

 

At a glance, it does seem like the two 105s are clearly ahead for capture of detail (and I don't mean sharpness).

 

Also, what are the exposure times?  

As mentioned in our Lens Sticky, you would want to find that the total EV per image was not more than 3EV over than that of the UV-Nikkor in order to say that a particular lens was truly "UV-capable". As Birna once noted, with enough time & patience you can force UV thru almost any lens. But long exposure times and forced UV will create some noise and a kind of muzzy look. (Sometimes total EV can be found in the EXIF which is useful when comparing a 3.5 to a 4.5. so that you don't have to make the EV calculations yourself. I think there are EV calculators somewhere. I don't seem to have one bookmarked.)

 

******

 

I will probably split the lens tests off into a separate topic. Lens tests are useful. But they might not be found under this topic.

 

*****

 

Yeah, you probably need a better subject. Do try some kind of medium range (under 20 feet) outdoor subject in strong sunlight.

 

Maybe look for some sunflowers in the grocery store. They are usually better subjects for such tests because of the contrast of the UV-dark center and the false yellow petal tips. Shoot them both outdoors (no flash) and indoors.

[ulf]

6 hours ago, Andrea B. said:

The first 9 images were made in ambient window light? That may not be enough UV light to bring out differences between "good" UV lenses and not-so-good ones if all are getting only light above 380 nm?

 

Yeah, you probably need a better subject. Do try some kind of medium range (under 20 feet) outdoor subject in strong sunlight.

 

Maybe look for some sunflowers in the grocery store. They are usually better subjects for such tests because of the contrast of the UV-dark center and the false yellow petal tips. Shoot them both outdoors (no flash) and indoors.

 

Both the light source and the subjects are important for evaluating how good a lens is transmitting for UV-photography.

 

The brightness of the UV-false blue flowers used above are an indication of the systems sensitivity at the very longest UV wavelengths.

(light source * flower reflectance * filter transmission * lens transmission * camera sensitivity)

Here there are not much indications of deeper UV-transmission as the blue flowers do not reflect those wavelengths.

 

Using flowers with UV-false yellow colours as Andrea suggests above is better.

 

---

An alternative approach that singles out the lens UV-transmission better is to use a stable 365nm LED torch as light source.

Then a UV-neutral matted surface is a suitable subject.

The surface should ideally be a virgin PTFE sheet that has been matted with a rough grit water-sand paper (220-300grit)

A rather matte anodised sheet of aluminium can work too.

 

For this setup it is extremely important that the distance between the torch and matted sheet is kept fixed.

 

The lenses tested should be set to the same aperture setting, maybe f/8

 

If there are a known lens candidate with reasonably good transmission, use that to find a good exposure time that gives a good exposure without any overexposure.

Use that exposure time for testing all lenses.

Shoot RAW and investigate the  the position of the peaks for different lenses compared to the best lens.

 

A lens with good deeper UV reach down to say 350nm will result in more saturated false UV colours, especially for yellow UV-false flowers and will lessen the demand somewhat of a really good IR cancelation of the UV pass filters.

[Andrew Dayer]

@Andrea B. & @ulf - thanks for your feedback and editing, Andrea. This was a post that got out of hand since it was only originally a quick-and-dirty exercise for myself but the timing of @DKoch's thread pushed me. I should of been clearer:
 

• Images are screengrabs of RAW files in Lightroom (hence unhelpful automatic file name). They seem much darker here than on my editing screen.
• All images illuminated with 365nm UV torch mounted in a fixed position (except last three flash images).
• Target area behind and left of the flowers (qualitative subject) is matt virgin PTFE. It is not Spectralon (yet!) but should be adequate for quick and dirty work
• Images were made at widest aperture (actually, to minimise fumbling with unfamiliar lenses ) at 1/60 and ISO800. Then corrected to ~EV 6.5 (ie f3.5 in this case)
• The Helios images were over corrected (darkened by 1 stop). I'll replace for the sake of form.

Thanks for the improvement suggestions. I'll see if there are sunflowers in the shops right now.

Outside - no go here in the UK in November. Not much UV, very wet and anything loose is immediately airborne!

I may do a limited re-shot. All though this was just a quick and dirty exercise, it was also a preliminary to assess useful wide angle UV transmitting lenses (insect and bird UV range) to screen trays of pinned insects.

[ulf]

3 hours ago, Andrew Dayer said:

 

• All images illuminated with 365nm UV torch mounted in a fixed position (except last three flash images).
• Target area behind and left of the flowers (qualitative subject) is matt virgin PTFE. It is not Spectralon (yet!) but should be adequate for quick and dirty work
• Images were made at widest aperture (actually, to minimise fumbling with unfamiliar lenses ) at 1/60 and ISO800. Then corrected to ~EV 6.5 (ie f3.5 in this case)
• The Helios images were over corrected (darkened by 1 stop). I'll replace for the sake of form.

 

It seams like you have all the ingredients to do my alternative suggested method and a sound setup.

 

With a 365nm torch those false UV-blue flowers are quite useless as test objects as they reflect very little of the 365nm light. The PTFE sheet would be enough.

 

If you can keep the aperture the same for all lenses you do not need to analyse any images or correct any exposure as long as you have the level reasonably right in the first place.

My suggestion would be to investigate the difference of the position of the peaks in the histogram directly.

 

I do not have the Lightroom so I do not know if it is a goos tool for that.

RawDigger or FastRawViwer are my tools for such comparisons. 

 

My reasoning behind the method is to separate different parameters as much as possible and find the pure transmittance.

Then you can tell how good any given lens is at the same aperture setting.

Most of the time it is good enough to have a reasonably good transmittance at 365nm.

[Lou Jost]

Remember also that transmittance % is helpful but needs to be taken together with the speed of the lens and how far it needs to be stopped down to achieve suitably sharp photos. Suppose you have a lens with 100% transmission at f/8 and a second lens with 50% transmission at f/8. If the second lens can be opened to f/4 without losing sharpness, but the first lens has to be at f/8 for good sharpness, then the lens with 50% transmission at f/8 will actually be much better (in terms of exposure time) than the the lens with 100% transmission at f/8.

 

 

[ulf]

@Lou JostThose are all good points that I too consider important.

 

Depending on what you want to photograph there are very different needs in a complex world of parameters.

That is why I recommend separating the parameters as much as possible instead of looking at the final imaging results.

It is the only way to fully understand what is happening here. 

 

Another situation to consider: If you are doing street or environmental UV-photography.

Then there are rarely objects that need a deeper UV-reach and sometimes you only want short exposure times and might want a short DOF.

Then a fast lens with more marginal UV-reach would be a good alternative even if the UV passed by the lens is not that strong.

The gain of a faster lens at a wider aperture might give the desired shorter exposure times.

In that case there will be a need for a better suppression of the secondary NIR-peak around 700nm as light in that region is not attenuated while the remaining UV-light around 400 nm will be lower due to losses in the fast lens

[DKoch]

here's my simple little test target... classic 365nm BLB fluoro, 410 nm deep violet and classic 470nm blue LED; their light is striking Kodak Neutral Gray cards.I'm not looking for deep UV at all ; I'm really going for 350nm and up into visible range....my thinking is that the blue 470nm LED is a visible wavelength/response reference.In this way I can compare the 410 and 365 nm signals to the 470 nm blue to gauge relative sensitivity. I'm using this simple target test on different camera systems ( with IR cutoff /OLPF removed); there are usually still some kinds of cover glass etc with unknown UV absorption.These are all high end digital cinema cameras; I also tested my unmodified Sony A7SIII... just for fun

 I'm using the Industar 50-2 as a constant lens...

[dabateman]

Well if you have one you can test a Tamron adaptall 17mm lens. It has decent upper UV transmission above 350nm.

https://www.ultravioletphotography.com/content/index.php?/topic/4521-the-quest-for-ever-wider-breaking-the-20-mm-barrier/&do=findComment&comment=44606

 

 

[DKoch]

why i just happen to own a Tamron 17mm adaptall!

[DKoch]

On 11/14/2023 at 7:53 PM, Lou Jost said:

Remember also that transmittance % is helpful but needs to be taken together with the speed of the lens and how far it needs to be stopped down to achieve suitably sharp photos. Suppose you have a lens with 100% transmission at f/8 and a second lens with 50% transmission at f/8. If the second lens can be opened to f/4 without losing sharpness, but the first lens has to be at f/8 for good sharpness, then the lens with 50% transmission at f/8 will actually be much better (in terms of exposure time) than the the lens with 100% transmission at f/8.

 

 

i just completed phase 2 of my Digital Cinema camera UV sensitivity testing ...well, the acquisition part at least! As i mentioned, I'm using the Industar 50-2 that was recommended to me, but also a very $$$ Zeiss Master Prime 50mm lens for comparison. One odd thing: one would expect such a high performance modern lens with (probably) 7 different types of optical glass and high tech coatings to be bullet proof in the near UV...at a glance, this may not entirely be so. Also, most of the recommended vintage lenses that have been recommended are an f3.5 usually. These Zeiss lenses I'm testing are T1.3 ( probably an f 1.1).In the visible spectrum they are great at that aperture.In theory they probably have almost 3 stops more aperture than the vintage lenses. It will be very interesting to see the results when the smoke clears and I get a chance to analyze the results carefully. I have one more family of cameras to test: the Arri Alexa; I'm getting a " clear" OLPF sent in from Arri New York to test. At worst, it might be a Shott BG 39 glass.

[Lou Jost]

15 minutes ago, DKoch said:

i just completed phase 2 of my Digital Cinema camera UV sensitivity testing ...well, the acquisition part at least! As i mentioned, I'm using the Industar 50-2 that was recommended to me, but also a very $$$ Zeiss Master Prime 50mm lens for comparison. One odd thing: one would expect such a high performance modern lens with (probably) 7 different types of optical glass and high tech coatings to be bullet proof in the near UV...at a glance, this may not entirely be so. Also, most of the recommended vintage lenses that have been recommended are an f3.5 usually. These Zeiss lenses I'm testing are T1.3 ( probably an f 1.1).In the visible spectrum they are great at that aperture.In theory they probably have almost 3 stops more aperture than the vintage lenses. It will be very interesting to see the results when the smoke clears and I get a chance to analyze the results carefully. I have one more family of cameras to test: the Arri Alexa; I'm getting a " clear" OLPF sent in from Arri New York to test. At worst, it might be a Shott BG 39 glass.

Sounds very exciting. I think that to keep prices reasonable, a high degree of optimization in the visual part of the spectrum may require some trade-offs in the correction in the "useless" part of the spectrum.

[DKoch]

On 11/17/2023 at 6:35 PM, DKoch said:

 

I've used these lovely Zeiss Master Prime lenses ( "super 35 / APS-C") a lot in IR work and I must say their performance is quite impressive in the "useless" IR wavelengths! Out at 950-1000 nm you could see increased softness...but we were shooting at f1.1-2.8 !!! It's possible though that this test might show a catastrophic drop in sharpness below 400nm. Right now I'm only looking at relative near UV response from camera to camera and eventually studying noise levels, especially in the blue channel.It remains to be seen how these miracle lenses perform in terms of sharpness in the near UV