Jump to content
UltravioletPhotography

A quick-and-dirty transmission test using a pinhole


OlDoinyo

Recommended Posts

I just took delivery of a Lens-Laser brand pinhole for my A-mount cameras, which enabled me to attempt something I have not been able to do to date: to test the UV bandwidth transmission of lenses using the color they exhibit when transmitting ultraviolet light. This method does not purport to be as rigorous or quantitative as transmission spectra or even Sparticle tests; however, a surprising amount can be learned by doing this. Only the limitations of the filter and sensor themselves constrain this method.

 

After taping a stepper ring to the pinhole cap to provide a filter thread, the Baader filter was attached and the pinhole assembly mounted on the A900. Useful exposure turned out to be about 10 seconds at ISO 3200 (the provider did not give the size of the pinhole.) Outdoors, the bottom of a stainless steel bin provided a neutral reference target. Display intent was BGR. The images are blurry as pinhole images tend to be; most were heavily cropped.

 

The target appeared thus through the pinhole. The historic window glass in the background looks quite orange:

 

post-66-0-71350100-1452485670.jpg

 

To establish the system performance, four filters were photographed against the target: a generic "UV Haze" filter, a Tamron skylight filter, a B+W 403, and a generic yellow (quasi-K2) filter:

 

post-66-0-03920100-1452486141.jpg

 

The moniker "UV-Haze" proved risible: as it turned out, this filter shows only a faint yellow tinge, and in fact outperformed almost everything else tested in terms of UV transmittance. The skylight filter shows the more typical bright orange hue indicating that only very long-wave UV is getting through. The 403, on the other hand, appears cyan, indicating that long-wave UV is being preferentially attenuated. The yellow filter appears black as expected.

 

As further reference, I went through my gear and chose four lenses I believed would be poor UV performers: two modern lenses and two older wide-angles from circa 1980 (which were easier to deal with because they do not default to minimum aperture when off the camera) and photographed those:

 

post-66-0-90473400-1452486808.jpg

 

I was not disappointed: most of these lenses show varying shades of dark orange when looking through their barrels, indicating that only very long wavelengths are getting through. The Zeiss 135, however, takes the biscuit: it does not seem to transmit any UV at all (though it defaulted to f/32 and the orifice was very tiny.)

 

Finally, I compared my lenses that I consider UV-capable. The lens on my Olympus 35RD was not available, as that camera is in the shop. There was a surprise among these results:

 

post-66-0-50794600-1452487215.jpg

 

The Katoptar shows a slight orange tint, which is a real shocker: this is a glassless optic, which is supposed to be rated down below 300 nm, and I am at a loss to explain this. The Asahi 35 and the Minolta Autocord's Rokkor-X show noticeably more orange tint, though nothing nearly as bad as in the previous set of examples. The Steinheil shows only a faint yellow tint, the only thing to outperform the so-called "UV-Haze" filter. I would assume that a true quartz optic would show no tint at all in this test, but none was to hand. There may be a way to extract more quantitative information from these images, but it is not clear if it would be worth the trouble.

Link to comment
A side comment is that I also found the Makowsky to work poorly in UV. I suspect the reflective coating on its mirror surfaces to be causing this unexpected result. (for those not familiar with the Makowsky, it's all mirrors no glass).
Link to comment

Clark, I will return to this and read carefully when I finish the current editing tasks.

Looks to be a terrific experiment!

 

I simply love the creativity of our UVP members.

Link to comment

Regarding Zerodur mirrors (the kind used in the Makowski,) I have no direct way of determining how those mirrors were sourced nor what might have been applied to them--but protective coatings on modern Zerodur mirrors are generally designed to interfere as little as possible with reflection (as there is no flare issue, it is unlikely that antireflection coatings would be used.) However, the choice of reflective metal might play a role. Aluminum mirrors have a reflectivity that decreases from over 90% at 400 nm to about 73% at ~330 nm before increasing again at lower wavelengths. This might be detectable in a test, but would probably not be much of an impediment to photography. Silver mirrors have a somewhat more severe rolloff, from about 93% at 400 nm to perhaps 2% at ~300 nm (before increasing again.) If silver mirror coatings were used, that might be more of a real practical problem--but I remain skeptical that any optic with such properties would be touted for use at 300 nm. The real problem with that optic, in any case, is the poor image sharpness, which is a challenge at all wavelengths.

 

Reference:http://www.newport.com/Zerodur-Broadband-Metallic-Mirrors/141088/1033/info.aspx

 

I have recently purchased a Tamron 21mm pursuant to recommendations on this forum; I will no doubt test it in a similar manner.

Link to comment
  • 3 weeks later...

Update: I have now tested the 35RD lens and the Tamron 21....

 

post-66-0-87975200-1453929489.jpg

 

This one does not exactly cover itself in glory, although it is not as bad as the worst examples...I would guess it goes down to about 350, maybe? It is not completely UV-incapable, as I have taken a number of b&w UV shots through it, some of which have been seen here. In color, it would probably do less well. Amusingly, one can see the UV attenuation in the viewfinder as well. The fact that this is a fast 6-element lens undoubtedly does it no favor in the UV.

 

post-66-0-62184500-1453929797.jpg

 

About the same, which is better than a lot of 21mm optics. I recall that the Sparticle test put the cutoff for this one in the 350-360 range. I don't have a filter for this one yet, so I cannot demonstrate images; however, I believe other members have done so elsewhere.

 

Note: I did not get the target intensity quite matched for these shots, so the upper one probably looks a bit worse than it is.

Link to comment

Nice follow up. I must try this too now that I finally found my pinhole cap.

 

So, camera + pinhole cap + BaaderU filter.

First shoot neutral target for later WB efforts.

Then place test lenses so the mount end faces pinhole.

Make shots, white balance them and see what you got.

 

I wonder if shining the 365nm UV-Led through the lens would make sense?

It's certainly more narrowband than sunlight (smile), but otoh would cut down on exposure length.

Link to comment
365 will only tell you about transmission at 365. Monochromatic sources will not give any color effect. Broadband illumination is needed for that.
Link to comment
  • 2 months later...

Postscript: Enna 28/3.5

 

A lens matching this description is reported in one of Dr Schmitt's tables as having anomalously good UV transmission for its focal length. Accordingly, I have procured and tested one:

 

post-66-0-15295100-1461272764.jpg

 

There seems to be a discrepancy here, and a bit of a disappointment. The color observed in the test photo is approximately the same as that observed for the Tamron 21mm lens, with a reported cut-on around 360 nm. Some version of this lens is reported as having a cut-on as low as 325 nm, but this lens does not appear to be an example of that--it is nowhere near in the league of any of the UV-capable optics in my possession other than the Tamron and maybe the 35RD's lens. Image quality is described in a separate post.

Link to comment
Regarding the Tamron 21/4.5: you did remove the tiny filter at the rear? Apparently this filter cuts off a lot of UV and in fact, on my specimen of the 21 mm, had acquired a distinct yellow hue (ageing?). Removing the filter has the nice side effect that near focus limit is significantly improved as well.
Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...