Jump to content
UltravioletPhotography

Approx. UV Transmittance of 17 un/underdocumented lenses (rough estimates)


SteveCampbell

Recommended Posts

SteveCampbell

Protocol: Mounted 5D mark II on tripod. Took photo of scene with Steinheil Cassar 2.8/50 + 77mm BG39/UG-11. Shot same scene with new lenses, aperture wide open, focusing with liveview, and switching back to Steinheil 2.8/50 every 5 lenses to ensure no change in lighting conditions. Compared photo histograms from each lens with Steinheil to find relative exposure difference. Corrected for maximum aperture difference.

 

NB: (1) Filter combination is imperfect. (2) Despite my Steinheil 2.8/50 control, natural lighting conditions change. (3) I rated to nearest 1/3 stop by my eye, so fine differences are not accounted for. (4) Overall this is a very simple and rough test, so no need to nitpick methodology [i know you guys want to! If any generous soul wants to send an Ocean Optics Vis-UV or sparticle board my way I can give you proper results ;) ].

 

[#] Lens name: -/+ transmission estimated by correcting for maximum aperture (-/+ absolute exposure difference)

 

In other words:

The first number is approximately how well the lens transmits ultraviolet light compared with the Steinheil 2.8/50. I.e. suitability for UV. More negative numbers indicate worse performance

 

The second number (in brackets) is the difference in shutter speed you would get for a given scene if you switched from the Steinheil 2.8/50 to the test lens. More negative numbers indicate slower shutter speeds vs the Steinheil 2.8/50.

 

[1*] Steinheil Cassar 2.8/50 camera lens: +0 (+0) <--- CONTROL LENS

[2] Carl Zeiss Jena Biometar 2.0/58 camera lens: -2 (-1)

[3] Triplet 2.8/78 projection lens: -1 (-1)

[4*] Pentacon AV 2.8/80 projection lens: +0 (+0)

[5] Triplet 2.8/100 projection lens: -1 1/3 (-1 1/3)

[6] Meyer Diaplan 2.8/100 projection lens (identical optics to trioplan 2.8/100): -2/3 (-2/3)

[7*] EL-Nikkor 5.6/80 enlarger lens: 0 (-2)

[8*] Steinheil Cassar 4.5/105 bellows lens: 0 (-1 1/3)

[9*] Neokino ~2.1/120 projection lens: -1/3 (+2/3)

[10*] Kodak 2.8/127 projection lens: -1/3 (-1/3)

[11*] EL-Nikkor 5.6/135 enlarger lens: -1/3 (-2 1/3)

[12*] Phot-All 3.5/135 bellows lens: -1/3 (-1)

[13*] Schneider Componar 4.5/135 enlarger lens: 0 (-1 1/3)

[14*] Pentacon AV 3.5/140 projection lens: -1/3 (-1)

[15] Triplet-6 2.8/150 projection lens: -4 (-4)

[16] Pentacon (Meyer Diaplan) 2.8/150 medium-format slide projection lens: -1 (-1)

[17] Triplet-1 3.5/150 projection lens: -1 1/3 (-2 1/3)

[18*] Wollensak 4.5/190 large-format enlarger lens: -2/3 (-2)

[19] Edar 3.5/200 projection lens: -1 1/3 (-2)

[20] Kodak 4.3/250 projection lens: -1 (-2 1/3)

 

Edit: [21*] Phot-All 3.5/135 bellows lens: +1/3 (-1/3)

 

In the list above I included an asterix [#*] next to lenses I thought to be notable. While the EL-Nikkors are well known and included as additional controls, there are a number of unfamiliar, apparently high-performing additions. Particularly exciting to me is the Neokino, a fast lens that appears to be Petzval in design and character (swirly bokeh, vignetting, central sharpness).

 

Enjoy!

 

Edit: I added the Phot-All after the fact for those still paying attention. It appears to at least match, and possibly beat the Steinheil for transmittance. I would speculate that since the Steinheil is so close to the maximum transmission recorded for a non-dedicated UV lens, it is possible that the Phot-All matches it, but appears superior due to an underreporting of maximum aperture. I believe it is a rebranded Lentar bellows lens - possibly the same as the Lentar that Andrea has tested previously. A curious design feature is a protruding rear element that extends backwards 3-4cm past it's mount ring.

Link to comment

Interesting selection of lenses, and good to see some of them are very close to the Cassar S.

 

Can I ask how you mounted all these projection and large format lenses - was it using Raf camera mounts? If so were any of them using the M65 to 67mm diameter adapter? If any of them did use that one, it would be interesting to know which ones as I have one of those adapters. I used it for something called a Pentax ZOCS lens, which to be honest did not perform very well for UV, and would be interesting to find something else that fits the adapter.

Link to comment
SteveCampbell

Interesting selection of lenses, and good to see some of them are very close to the Cassar S.

 

Can I ask how you mounted all these projection and large format lenses - was it using Raf camera mounts? If so were any of them using the M65 to 67mm diameter adapter? If any of them did use that one, it would be interesting to know which ones as I have one of those adapters. I used it for something called a Pentax ZOCS lens, which to be honest did not perform very well for UV, and would be interesting to find something else that fits the adapter.

 

Exactly, I looked preferentially for very old triplet or 4-element lenses.

 

For the projection lenses I used a series of RafCamera adapters:

M52 to 42mm

M52 to 52.5mm

M65 to 52.5mm

M65 to 62.5mm

Used with various M42, M52, M65 helicoids.

 

For the enlargers I used different custom solutions for each. For the Wollensak 4.5/190 I used pliers to force it's metal mount ring into a 67mm to 72mm step up ring, then used a series of sequentially smaller step-up rings until I reached 39mm, at which point I mounted it to a exa tilt-shift bellows by means of an exa to M39 adapter, and mounted the bellows to the camera with an EF to exa adapter.

 

For the componar, I disassembled an A Schact München 3.5/135, and screwed the rear metal segment onto the threads of the componar, creating an exa mount that could be directly adapted to the bellows mentioned above.

 

For the Phot-all I lined up the screws of the M52-to-45mm with the projections of the bayonet mount, and used a 42-to-52mm step-up ring to adapt to a short M42 helicoid.

Link to comment

Wow, very cool, thanks for the information Steve. Raf doesn't make an adjustable iris for his adapters, but I did come across an M42 one on ebay while I was looking, if that is of use;

 

https://www.ebay.co.uk/itm/Adjustable-Iris-aperture-diaphragm-2-44mm-w-M42x1-Thread-For-Camera/323327078688?hash=item4b47cbb120:g:6M8AAOSwBLlVaRzX

 

I've not tried them myself, but could be interesting for some of the projection lenses.

Link to comment
SteveCampbell

Wow, very cool, thanks for the information Steve. Raf doesn't make an adjustable iris for his adapters, but I did come across an M42 one on ebay while I was looking, if that is of use;

 

https://www.ebay.co.uk/itm/Adjustable-Iris-aperture-diaphragm-2-44mm-w-M42x1-Thread-For-Camera/323327078688?hash=item4b47cbb120:g:6M8AAOSwBLlVaRzX

 

I've not tried them myself, but could be interesting for some of the projection lenses.

 

That aperture is a fantastic idea ... I would probably need a M42 to M65 step-up to use it. It would pose some trouble for many of the projection lenses however, as many need to be mounted very close to the sensor. More doable for mirrorless cameras I'd imagine.

Link to comment

Does a negative mean better than the Steinheil or worse? As in is your #15 4 stops slower than the Steinheil or 4 stops better?

I am assuming it means worse. For tests like this I like to use constant lighting, using 365nm led bulbs or a compact fluorescent black light.

But still you have interesting set of results.

I feel I have blown enough money on lenses though. Unless there ends up being something really fast. Oddly I seem to like the Sigma 30mm f2.8 art over my Pentax 85mm f4.5 UAT. Once I get my EM1 conversion done, that might change though. As my stock EM1 can see down to 370nm. I am not sure how deep the Sigma goes.

If you have a mirrorless Sony or M43rds its an amazing UV autofocus lens.

Link to comment
SteveCampbell

Does a negative mean better than the Steinheil or worse? As in is your #15 4 stops slower than the Steinheil or 4 stops better?

I am assuming it means worse. For tests like this I like to use constant lighting, using 365nm led bulbs or a compact fluorescent black light.

But still you have interesting set of results.

I feel I have blown enough money on lenses though. Unless there ends up being something really fast. Oddly I seem to like the Sigma 30mm f2.8 art over my Pentax 85mm f4.5 UAT. Once I get my EM1 conversion done, that might change though. As my stock EM1 can see down to 370nm. I am not sure how deep the Sigma goes.

If you have a mirrorless Sony or M43rds its an amazing UV autofocus lens.

 

Yes indeed, negative means worse transmittance / slower shutter speed. Was on the fence about how to present it +/-. #15 is indeed an abysmal UV performance, which is unfortunate as it renders beautifully soft and glowing artistic images in visible light.

 

I've previously posted tests using a 365nm source, but it doesn't account for shorter wavelengths where higher UV performance lenses differentiate themselves, and has a fair bit of longer UV contamination due to lightsource lens fluorescence (this isn't accounted for in LED spectral distribution curves since they measure directly from the diode). I've found that the 395nm/blacklight sources don't give much information since even modern lenses pass a fair bit at 395. Since I photograph my subjects under natural light anyway, it made sense to use the sun.

 

I made a post a while back where I suggested that the 365nm to 395nm transmission ratio could be a proxy for depth of transmission of a particular lens (the idea probably has merit), but I had a fair number of responses suggesting that the "sparticle board" method is entirely superior (which is quite true indeed).

 

Edit: I modified my description above to clarify interpretation of the negative values.

Link to comment

Yes the plastic dome on some Led bulbs does shift the output to be a little more broad spectrum. I cut the plastic dome off my 365nm Leds bulbs and its only a little better. These bulbs I speak of are different than the 405nm black Leds I posted about before. I got the Lixada 365nm bulb from Amazon.

I also now have a Reptile mercury vapour lamp that I still need to test as well. It was reported to have great UVb lines, so hopefully will be better for lower uv bands. The glass on the front is clear. But it does get hot, which is why I have liked the LEDs. I built a special housing for it.

Anyway, good test if it helps you decide on the best lenses. I was surprised that my pentacon six lenses mostly all were bad for uv. The single coated 50mm f4 was just as bad as my multi coated 50mm. Neither pass much uv or even at 405nm. I still haven't tested my 180mm f2.8 Olympic lens though. The pentacon six 80mm f2.8 is about 2 stops slower than my Steinheil.

Link to comment
SteveCampbell

That's funny, I was actually looking at the same reptile lamp a while back. Yes, the Pentagon Six 50, 80, and 180 are all poor performers by my experience - too much glass, too many elements, too many coated surfaces. But fantastic for other spectra when mounted on a tilt-shift adapter, and all together at a tiny fraction of the price of a modern tilt-shift.

 

Link to comment
  • 4 weeks later...

Hi, Steve, very interesting test you did. One question, did you use the same aperture setting to get the second number in the brackets? I saw "EL-Nikkor 5.6/80 enlarger lens: 0 (-2)", looks like -2 accounts for the 2 stop difference between f5.6 and f2.8? I once compare both lens by shooting a same UV landscape photo, I used f5.6 on both lens and found EL-Nikkor to be 1/3 of a stop faster than Cassar S.

 

I was also looking for potential performer among triplet lenses a while back, and I found that a lot of the older rangefinders use simple designed lenses. One of my finding was the 40/2.8 lens made by Seiko on a Mamiya rangefinder, a link of its photo here:http://3.bp.blogspot...s1600/rank2.jpg

I found this lens to have the same exposure as Cassar S at any same aperture, and its centre sharpness and contrast is only better. The thing I'm not happy about the Cassar S has always been its low contrast, not sure is it just a problem of my copy. However, taking these rangefinder cameras apart to get a lens was a real trouble, some more I had to adapt it to M42 screw mount of my helicoid.

Link to comment

Hi, Steve, very interesting test you did. One question, did you use the same aperture setting to get the second number in the brackets? I saw "EL-Nikkor 5.6/80 enlarger lens: 0 (-2)", looks like -2 accounts for the 2 stop difference between f5.6 and f2.8? I once compare both lens by shooting a same UV landscape photo, I used f5.6 on both lens and found EL-Nikkor to be 1/3 of a stop faster than Cassar S.

 

I was also looking for potential performer among triplet lenses a while back, and I found that a lot of the older rangefinders use simple designed lenses. One of my finding was the 40/2.8 lens made by Seiko on a Mamiya rangefinder, a link of its photo here:http://3.bp.blogspot...s1600/rank2.jpg

I found this lens to have the same exposure as Cassar S at any same aperture, and its centre sharpness and contrast is only better. The thing I'm not happy about the Cassar S has always been its low contrast, not sure is it just a problem of my copy. However, taking these rangefinder cameras apart to get a lens was a real trouble, some more I had to adapt it to M42 screw mount of my helicoid.

 

Hi Bruce,

 

I used each len's maximum aperture for the number in brackets, and corrected for the difference in maximum aperture to arrive at the first number. If we're coming to within 1/3rd of each other in such tests, it's within the realm of variability I would expect - especially considering that I rounded to the nearest 1/3rd.

 

Yes, I would imagine some of the old rangefinder lenses to be good, not just because of the small number of elements and weaker coatings, but also because of the lower absolute amount of glass facilitated by such a small flange distance. Unfortunately that flange distance also prevents me from experimenting with them at the moment as I'm using an old converted Canon 5Dmk2 with its huge dSLR flange distance

 

I believe my 5.6/80 may suffer from some haze, as it displayed poor contrast. Perhaps the same is true of your Steinheil? If so, between out two tests, a hazed 5.6/80 + hazed 2.8/50 could certain add up to an exposure discrepancy of 1/3rd stop.

Link to comment

Hi Bruce,

 

I used each len's maximum aperture for the number in brackets, and corrected for the difference in maximum aperture to arrive at the first number. If we're coming to within 1/3rd of each other in such tests, it's within the realm of variability I would expect - especially considering that I rounded to the nearest 1/3rd.

 

Yes, I would imagine some of the old rangefinder lenses to be good, not just because of the small number of elements and weaker coatings, but also because of the lower absolute amount of glass facilitated by such a small flange distance. Unfortunately that flange distance also prevents me from experimenting with them at the moment as I'm using an old converted Canon 5Dmk2 with its huge dSLR flange distance

 

I believe my 5.6/80 may suffer from some haze, as it displayed poor contrast. Perhaps the same is true of your Steinheil? If so, between out two tests, a hazed 5.6/80 + hazed 2.8/50 could certain add up to an exposure discrepancy of 1/3rd stop.

Ya, probably, maybe I didn't get a good copy of Steinheil, however I'm most happy about the image quality of my El-Nikkor, especially its good contrast and corner to corner sharpness.

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...