Jump to content
UltravioletPhotography

Optimizing a lens for UV transmission


lost cat

Recommended Posts

I've been wondering about the possibility of optimizing an older camera lens to transmit UV light. Ive read some people have had some success scrubbing away the coatings on Nikon E lenses while others say this has a high risk of grinding the glass down.

 

My question is whether anyone here has successfully remove coatings chemically or thermally. Many years ago I worked at a company that developed one of the first commercial capillary electrophoresis (CE) instruments for DNA sequencing. The capillary in CE is a tiny hollow fused-silica tube coated on the outside with polyimide to give it mechanical strength while the inside was coated in-house with an organic coating chemically bonded to the silica to prevent electroosmotic flow. Occasionally we had a problem with fluorescent dyes contaminating the internal coating so we developed a protocol to remove the coating. We tried a variety of acids and bases and found the only thing that worked was a 0.1M hydroxide (NaOH or KOH), 50 alcohol/50 water solution. There was some danger of etching the internal surface if the solution went above 1M and/or if the solution was left in too long but otherwise it worked beautifully. Its also a common solution to clean laboratory borosilicate glassware of most gunk that otherwise just won't come off.

 

A small portion of polyimide coating needed to be removed to make an optical window. AFAIK the cleaning solution described above did not work for the polyimide coating; well perhaps it would have worked but there were quicker ways to do the job. The coating could be removed in a second or so by applying heat from a flame or an electrical heating element (like a toaster). We also used hot concentrated (18M) sulfuric acid. This was considered to yield superior optical results but not by much.

 

So has anyone tried to remove lens coatings by either soaking the lens in an alcohol hydroxide solution or heating the lens to the point where the coating flaked off? Heating has risks but if the temperature changes are gradual enough the lens should not crack, nor should the temperature be so hot as to deform the lens.

 

There is also the issue of the adhesives. Canadian balsam used in older lenses is supposedly easy to removed by soaking in hot water (again let the water and lens cool slowly to prevent cracking). Are there any modern UV transparent adhesives that can be used to put the lenses back together? Most such modern adhesives are cured with UV light so these will of course not work.

Link to comment

Greetings, Lost Cat, and welcome to UVP by the way!

 

I've often wondered along similar inquisitive lines, such as the ones which you've presented herein. Especially since I happened to have tinkered with virtually thousands of vintage lenses of all makes and designs, in order to search for potential UV performers.

 

Although I have no experience with de-gluing and then re-gluing (with UV-transparent optical cement) any optical formulas which involve cemented element groupings, ultimately I have concluded that there is no point of even bothering with such an added step, since there are many existing "accidental" UV-capable lenses out there that are based upon air-spaced optical formulas with no cementing. Thus, in my honest opinion, an unnecessary and frivolous endeavor, in the end, given the fact that it can be avoided altogether in preference to optical designs that are cementing-free.

 

(One of these days, I plan to publish my cumulative findings of discovering hundreds of lenses that transmit more than sufficient levels of UV, and no cementing modification is ever required for those lenses).

 

As for coating removal, that is a completely different beast, all its own, of course. And so far, I have had some limited success with a two-part process. One that first involves a 24-hour white-vinegar soak of entire elements (those that are coated), followed by a low-abrassion glass-polishing treatment using an extremely fine optical-grade form of cerium oxide powder known as 'Rocket 444', which was precisely designed for just that. This has worked for me, at least with older, weaker, single-coated glass. But as for more resilient, multi-coated glass, I probably wouldn't even bother with it.

 

Oh, and in case anyone ever wondered: The acid in white vinegar will NOT eat away at the glass surface, because this weak acid does not interract with glass (whereas some other acids do). At least, nothing of note, when limited to a 24-hour soak. But it will loosen the bonds of older, weaker coatings quite nicely ... which then makes the second step (polishing the coating away with 'Rocket 444'-brand cerium oxide, via a dremel tool and a soft felt-based buffing wheel attachment) much easier.

 

Then again, the coatings which I have successfully stripped away were probably never that UV-suppressive to begin with, since I mostly deal with vintage glass which came before the proprietary multi-coating age.

 

(Do note: Rocket-444 powder first has to be mixed with water, preferentially distilled, in a 1:1 mixture ratio, to turn it into a paste ... before applying it to the polishing wheel.)

Link to comment

Oh, one last thing: I cannot guarantee that my process described above will work for ALL attempts ... but it has had limited success with some types of coatings. At the very least, it has thinned them down, to a notable degree. And in best-case scenarios, it has removed them completely.

 

EDIT: Go easy on your dremel's RPM speed setting. Start slow, and work your way up, cautiously. Even with the added "cushioning" created by the exceptionally fine-grit Rocket 444 polishing paste between the felt-based buffing wheel and the glass, a high-enough RPM (and excessive pressure applied by the hand) may still risk ripping some glass away, if one is too forceful and impatient.

 

I would also suggest first weighing the "pros" and "cons" of wanting to remove coatings from any given lens, to begin with. Especially if the lens already transmits sufficient UV, as it is. Because the thing is, while removing a coating MIGHT squeeze out yet a higher transmission of UV, it can also rob the lens of its contrast and resistance to flaring. Thus, if I find a lens to ALREADY transmit enough UV to be of satisfactory use, I'd rather keep the coating and retain enhanced contrast and/or flaring resistance, than get greedy for yet more UV throughput but turn my contrast to crap. :-)

Link to comment

We already have 120+ UV-capable lenses listed in the UV Sticky

I'm sure there are more.

 

I'm still liking my UV-Nikkor the best, however. It has good optical qualities in addition to its UV-ness. :D

But all those nice 35/3.5 under various labels are lots of fun and a couple of them very good.

 

Many lenses have inner elements which are coated. Removal of coatings could get very labor intensive!

Link to comment

Hi Iggy,

 

Yes I'm looking forward to seeing that list! I don't suppose you know of any *cheap* UV capable zoom lenses :rolleyes:

 

@ Andrea

Unfortunately a UV-Nikkor or any other such speciaty lens is a bit over budget so I need to make do with cheap *accidental* lenses - thank you for the UV sticky! I hope to see it expanded as new lenses are discovered.

 

As I mention I have come across a few websites which discuss the removal of coatings to increase UV transmission so there is some interest in this question. As you point out if the inner elements may also be coated making this a time consuming and risky project. If however the coatings can be removed by just soaking the elements in a stripping solution for a few minutes without damaging the glass such a project is much more feasible. Since alcoholic hdroxide is a well known method for cleaning filthy lab glassware I was curious if anyone had tried it on lens elements. Whether its worth doing, well that is another question altogether.

Link to comment

Just tell her that these lenses and your fascinating UV "hobby" keep you out of other kinds of trouble. :D

 

I'm sorry we do not have more info on Canon lenses. By historical accident, so to speak, most of the early digital UV photography was done using Nikons because Bjørn has always used them. So most of our information has accrued around the Nikon brand. Also Nikon never changed their mount, so all their old lenses have been easily mounted on modern digicams to be UV tested. Then too, Nikon was one of the few manufacturers of a dedicated UV lens, the UV-Nikkor (now the UV-Rayfact). So it would seem that until the arrival of mirrorless, the UV drift was generally in the Nikon direction with a few side trips into the F-mount Fuji S line.

Anyway.....anything you might find out about the UV capability of Canon lenses would be a most welcome addition to the Lens Sticky.

 

My Panasonic 14-42, original model, is, surprisingly, fairly UV friendly. It's for m4/3 sensor size. So the possibility certainly exists that there are some other UV friendly zooms.

 

You might possibly want to pick up a used mirrorless m4/3 body to convert from the Panasonic Lumix G line, for example, because this would permit you to make use of a wider range of UV-capable lenses after acquiring a couple of mount adaptors. Those Gs really make nifty conversions. LIttle gems. And if you get a used one from KEH (or similar reputable used dealer), you can get it for cheaps and not upset the wife.

Link to comment

I had [limited] success with the Nikon 36-72 mm f/3.5 SE zoom lens for UV work. The SE line had much simpler coatings than their contemporary Nikkor siblings. The basic issue with the zooms is their multitude of optical elements thus even though each element transmits some or even a lot of UV, the entire assembly quite efficiently attenuates UV in the end.

 

Sometimes I fear people are obsessed by a a single criterion, viz. how far into UV a lens can transmit. In order to document most UV-related phenomena in nature, having adequate UV transmittance to say 350 nm suffices. However, the final usefulness of a lens for UV work depends to a large extent on its optical performance sharpness, contrast, CA artefacts) when pushed into the UV range. Here many potential candidate lenses fail miserably as they need a lot of stopping down in order to attain acceptable sharpness, and many never get really sharp at all due to low image contrast.

 

Flare caused by the lens coatings being inadequate in UV is widely seen and the smallest visible light leak in the setup through viewfinders, adapters or filter holders can completely destroy the final UV capture.

 

Focus shift is another issue that can be very troublesome, for pure UV as well as for multispectral work. It is true LiveView or mirrorless cameras can circumvent focus shift issues - that is, when there is sufficient UV present so the electronic finder shows anything at all other than coarse noise. Indoors one might need a powerful UV-emitting torch as a focusing aid and outdoors the UV levels can be so low that the focusing aids fail. I encounter the latter a lot for UV video with my Panasonics when shooting on dull or rainy days.

Link to comment

Canon claims their high end lens multicoating transmits UV:

 

http://www.canon.com...ght/003/03.html

 

Can anyone verify if there are any Canon lenses good for UV?

 

None of the Canon lenses tried by me over the years did anything nearly acceptable in UV. Doesn't mean there cannot be potential useful lenses out there of course, but so far they haven't shown up on my radar.

 

To the best of my knowledge, Canon never made anything even remotely similar to the UV-Nikkor. A bit strange perhaps as their use of fluorite elements goes way back in time. Not certain they did quartz elements though.

 

The Canon cameras, once properly modified, should be capable of UV photography. One can adapt any of the "35/3.5" class lenses, get a Costal Optics 60/4 in Canon mount (I believe this version exists, please correct me if not), or use a lens adapter to mount a UV-Nikkor.

Link to comment

Just tell her that these lenses and your fascinating UV "hobby" keep you out of other kinds of trouble. :D

 

I'm sorry we do not have more info on Canon lenses. By historical accident, so to speak, most of the early digital UV photography was done using Nikons because Bjørn has always used them. So most of our information has accrued around the Nikon brand. Also Nikon never changed their mount, so all their old lenses have been easily mounted on modern digicams to be UV tested. Then too, Nikon was one of the few manufacturers of a dedicated UV lens, the UV-Nikkor (now the UV-Rayfact). So it would seem that until the arrival of mirrorless, the UV drift was generally in the Nikon direction with a few side trips into the F-mount Fuji S line.

Anyway.....anything you might find out about the UV capability of Canon lenses would be a most welcome addition to the Lens Sticky.

 

My Panasonic 14-42, original model, is, surprisingly, fairly UV friendly. It's for m4/3 sensor size. So the possibility certainly exists that there are some other UV friendly zooms.

 

You might possibly want to pick up a used mirrorless m4/3 body to convert from the Panasonic Lumix G line, for example, because this would permit you to make use of a wider range of UV-capable lenses after acquiring a couple of mount adaptors. Those Gs really make nifty conversions. LIttle gems. And if you get a used one from KEH (or similar reputable used dealer), you can get it for cheaps and not upset the wife.

 

 

Already done! I purchased a gently used Panasonic G5 a few weeks ago with the intention to convert it. Good to know the 14-42 lens it came with is UV friendly. I have a used 1.25" Astrodon Sloan U filter on the way so I'm looking forward to trying it out :)

Link to comment

No, mine was the original 14-42. This zoom model was changed somewhere along the line. So I do not know whether later models have the same UV friendliness. At any rate, remember that you can force UV through almost any lens with long enough exposures. Which is not to say that you will get a particularly good UV foto. Or be able to record much below 380nm. :D

 

Anyway, the G5 should be nice.

Do you know whether that Astrodon SloanU needs an IR blocker? We will be looking forward to seeing some photos from it.

Link to comment

There is also the issue of the adhesives. Canadian balsam used in older lenses is supposedly easy to removed by soaking in hot water (again let the water and lens cool slowly to prevent cracking). Are there any modern UV transparent adhesives that can be used to put the lenses back together? Most such modern adhesives are cured with UV light so these will of course not work.

 

Why would you want to de-cement lenses that use Canada Balsam?

Link to comment

Already done! I purchased a gently used Panasonic G5 a few weeks ago with the intention to convert it.

 

I have two G5's converted to full-spectrum, and could not be happier with them, for the majority of my UV-photographic needs. I've seen someone convert a G6, but there really isn't too much of an improvement in the G6 model, versus the G5 model, to warrant the extra difference in price. If someone wants a significant-enough improvement from the G-series, then they should consider moving up to the GH-series. Otherwise, I'd stick with the G5 as the all-around best G-series camera to convert to full-spectrum use.

 

(This, of course, if you choose to remain with Panasonic. However, the Olympus PEN / OM-series mirrorless and Sony E-mount / NEX-series mirrorless cameras are also good candidates for full-spectrum conversion).

 

Can anyone verify if there are any Canon lenses good for UV?

 

I haven't gotten around to testing any Canon lenses for UV transmission, mostly because of some of the points that Andrea already brought up. Chiefly, that because Canon has changed its camera mounts quite a number of times since its "breech-lock" mounts (FL / FD), then any current adaptations to other non-native camera systems can be a nuisance of sorts. Especially when adapting to cameras with larger FFD's (flange focal distance), which then requires an adapter that introduces an extra piece of glass (to restore infinity-focus abilities) ... and this corrective glass inside of the adapter can degrade image quality considerably. Especially for UV-work, if the glass is of an inferior design.

 

Whereas, there are two traditional (mirror-based) DSLR's with large FFD's ... particularly Pentax and Nikon ... which have retained their mount designs all of these years (even from the beginning of their creation), so that even with the addition of electronic contacts over the years, these mounts still remain virtually "universally" backwards-compatible with all "legacy lenses" for their respective mounts. (Ex: All Pentax K and M42 screw mount lenses can still be mounted on the most recent Pentax KA / KAF camera mounts without any issues except for the dreaded "Ricoh pin" on some lenses which can be easily removed, and likewise, all older Nikon F lenses can be mounted to the most recent Nikon AI systems, with a few exceptions and caveats).

 

(Not to mention, I consider the Canon FL / FD breech-lock mount to be an epic failure, on Canon's behalf. In their attempt to make a mount overly complex, they actually made it more prone to mechanical failure. It's just a horrible idea, which is probably why they eventually abandoned it and moved on to a "three-lug" bayonet-type mount, like just about every other camera mount out there.)

 

So, this is why I have essentially avoided even trying to check with Canon lenses as potential UV-performers, because I just don't need the added hassle of lens-mounting issues / modifications / adaptations.

 

As for mirrorless cameras with much smaller FFD's ... well ... this is a whole different "ballgame", now, so to speak. Which is why mirrorless cameras have become increasingly more popular over time, especially for UV-conversion work, given the fact that the much smalller FFD's make them even more universally-compatible with a broader range of lenses (without the need of corrective glass to retain infinity-focus, that is).

 

There are many other advantages of adapting a mirrorless, EVF-based system for UV photography ... over a traditional flapping-mirror DSLR. But that would go beyond the scope of this thread's main topic, so it's best saved for a different discussion. B)

Link to comment

Yes I'm looking forward to seeing that list! I don't suppose you know of any *cheap* UV capable zoom lenses :rolleyes:

 

I've essentially and almost exclusively focused the majority of my UV research into prime lens. So, I am sorry to say that I have no experience with testing zooms for UV capability. However, I have had a "hunch" for a few that I have in my possession, but have not gotten around to that yet. Perhaps some day ... (as if I don't have enough "project ideas" on the back-burner as it is. :unsure: Haha.)

Link to comment

But all those nice 35/3.5 under various labels are lots of fun and a couple of them very good.

 

I must say, though, that this past year of my ongoing research and testing of various prime lenses of all makes and models has been quite illuminating (no pun intended) ... because I have since discovered many other wider-angle primes that can easily compete with the hoard of 35mm F/3.5's out there, when it comes to UV transmission.

 

Some of the others that I have found, that can hold their own:

 

Specific models of 35mm F/2.8, 35mm F/2.5, 30mm F/4, 28mm F/3.5, 28mm F/3, 28mm F/2.8, 28mm F/2.5, 25mm F/3.5

 

All of them, just as capable (if not more so) as the now-famous Kyoei / Kuribayashi 35mm F/3.5 "clones" out there. (Meaning, the ones I am talking about can transmit down to 330nm-ish, at minimum.)

Link to comment

There is also the issue of the adhesives. Canadian balsam used in older lenses is supposedly easy to removed by soaking in hot water (again let the water and lens cool slowly to prevent cracking). Are there any modern UV transparent adhesives that can be used to put the lenses back together? Most such modern adhesives are cured with UV light so these will of course not work.

 

As far as adhesives, being a UV curable adhesive does not exclude it from being a UV transmitting adhesive, some are designed specifically for UV transmission.

I have tested and compared all of these:

Norland 63 or 88 adhesives can be used for UV, these are one part UV curable adhesives (they harden when exposed to UV light). I have seen no differences between 63 and 88 in my tests.

Norland Adhesives

Epotek 305 is a two part epoxy type adhesive which transmits UV even lower, down to 250nm, no UV needed for hardening, but much more difficult to work with than Norland.

Epotek 350

 

My favorite UV lens is the Kuribayashi 35mm f/3.5. It is the Petri version of the Kyoei 35mm f/3.5, and there are other Kyoei clones of this.

I will post about it separately as not to fill up this topic with bulk.

Here:

http://www.ultraviol...v-transmission/

Link to comment

Specific models of 35mm F/2.8, 35mm F/2.5, 30mm F/4, 28mm F/3.5, 28mm F/3, 28mm F/2.8, 28mm F/2.5, 25mm F/3.5

 

All of them, just as capable (if not more so) as the now-famous Kyoei / Kuribayashi 35mm F/3.5 "clones" out there. (Meaning, the ones I am talking about can transmit down to 330nm-ish, at minimum.)

 

Igor, you do realize that we all now want to see some proof, pictures, etc...

Link to comment

I don't suppose you know of any *cheap* UV capable zoom lenses :rolleyes:

 

Why do you need a zoom?

 

I would recommend you this:

Pick one, just one lens with a focal length equivalent to full frame 35mm-105mm and shoot with it for a month.

Put it on your favorite camera and use that lens for a month and take pictures every day, of every possible subject.

And do not use any other lens or camera. One lens, one camera, one month. See what happens...

Link to comment

Igor, you do realize that we all now want to see some proof, pictures, etc...

 

Of course, Alex. I would expect no less from the UV community. Which is precisely why it has been such a long and drawn-out process to have every single one of the lenses tested and re-tested extensively (for every criteria, including transmission curves, sharpness, distortion, etc.), and get all of the data uniformly organized before I finally make a formal publishing of it. I take these findings very seriously, so until I am ready to release this information with reliable and complete data, then this is why I have held off.

 

This has been no easy task, to be sure, since my resources and budget have been limited, my free time also limited (I have to make time for this ongoing research in between my full-time work), and other hurdles that I have had to overcome.

Link to comment

 

 

Why would you want to de-cement lenses that use Canada Balsam?

 

To repair a seperating element and/or remove fungus and hopefully improve UV transparency along the way.

 

 

 

 

 

 

 

Link to comment

 

 

As far as adhesives, being a UV curable adhesive does not exclude it from being a UV transmitting adhesive, some are designed specifically for UV transmission.

I have tested and compared all of these:

Norland 63 or 88 adhesives can be used for UV, these are one part UV curable adhesives (they harden when exposed to UV light). I have seen no differences between 63 and 88 in my tests.

http://www.norlandpr...siveindex2.html

Epotek 305 is a two part epoxy type adhesive which transmits UV even lower, down to 250nm, self hardening, no UV needed, but much harder to work with than Norland.

http://www.epotek.co...detail.html?cid[0]=3

 

My favorite UV lens is the Kuribayashi 35mm f/3.5. It is the Petri version of the Kyoei 35mm f/3.5, and there are other Kyoei clones of this.

I will post about it separately as not to fill up this topic with bulk.

Here:

http://www.ultraviol...v-transmission/

 

Very interesting, thank you for the tip!

Link to comment

To repair a seperating element and/or remove fungus and hopefully improve UV transparency along the way.

 

According to my references, Canada Balsam transmits UV very well.

Link to comment

According to my references, Canada Balsam transmits UV very well.

 

I suppose that may be wavelength dependent:

 

"Absorption of light by canada balsam restricts the use of Nicol prisms to the ultraviolet to about 0.35um. To extend the use of the polarizing prisms to the ultraviolet, Canada balsam should be replaced by air or UV transmitting oils."

 

Kailash K. Sharma Optics: Principles and Applications pg 72

 

https://books.google.com/books?id=d8QU7tbzLKwC&pg=PA72&lpg=PA72&dq=Canada+Balsam+uv+absorb&source=bl&ots=_uuQU4TcLF&sig=-JHh0CgQACXsL9Qc5ic601HedoY&hl=en&sa=X&ved=0ahUKEwjkw8Cjg7fJAhUMM4gKHfAGBnoQ6AEINDAE#v=onepage&q=Canada%20Balsam%20uv%20absorb&f=false

 

If the elements of the achromat are made from crown and flint glass, themselves with an inherent cutoff at about 350nm, then it seems replacing good Canada balsam may be counterproductive after all. If however replacement is done to replace damage then it is good to know the balsam won't be the weak link.

 

The ability to fix a mistake with balsam is a nice feature too

 

Update: Another reference claims Canada balsam has a cutoff of 50% at 300 um. Also lists the 50% cutoff for cellulose caprate at 240 nm.

 

Paul R. Yoder, Jr. Opto-Mechanical Systems Design, Third Edition pg 131

 

Norland 61 is a no-go for UV:

 

http://jimshomeplanet.com/lensglue.html

 

Some references that may be useful:

 

Pellicori, S.F., "Optical Bonding Agents for IR and UV Refracting Elements" Proc. SPIE 1535 Passive Materials for Optical Elements, 48 (Nov 1, 1991) doi:10.1117/12.48302

 

http://proceedings.spiedigitallibrary.org/proceeding.aspx?articleid=968944

 

Pellicori, S.F., "Transmittances of Some Optical Materials for Use Between 1900 and 3400 Angstroms" Applied Optics, V3, I3, pp 361-366 (1964)

Link to comment

"Norland 61 is a no-go for UV"

That is correct, however Norland 63 and 88 will work.

 

"The ability to fix a mistake with balsam is a nice feature too"

It is possible to fix mistakes with 63 and 88. Using pre-cure and cure stages benefits the process in several ways. When all else fails, the adhesive can be disolved.

Read the info Norland has about working with products.

Link to comment

"Norland 61 is a no-go for UV"

That is correct, however Norland 63 and 88 will work.

 

"The ability to fix a mistake with balsam is a nice feature too"

It is possible to fix mistakes with 63 and 88. Using pre-cure and cure stages benefits the process in several ways. When all else fails, the adhesive can be disolved.

Read the info Norland has about working with products.

 

Yes they can by soaking overnight in methylene chloride. Some people may not feel comfortable working with it:

 

https://www.osha.gov/Publications/osha3144.html

 

Others like myself who used it as a damn good tape head cleaner back when there were tape heads to clean wouldn't think twice about using it :D

 

Good news is both these adhesives are almost 100% transparent down to 350nm.

 

https://www.norlandprod.com/adhesives/noa%2063.html

 

https://www.norlandprod.com/adhesives/noa%2088.html

 

Looks like 88 is the better choice for the lower 300s but the glass itself will probably be the limiting factor for either.

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...