Lens transmission in the UV - latest update

[Andy Perrin]

Remember these are transmission graphs, all done at maximum aperture (this value should not be aperture dependent).

True that -- and it suggests a test: what happens if you change the aperture? At what point do you start to see a difference? At the very least, for a nearly closed lens, the noise should be pretty bad.

[dabateman]

Andy for a spectrometer test it may not be useful at all to drop the aperture. The light beam diameter, I am assuming is quite small, relative to even an F8 aperture opening.

You would see a difference in photos.

Most interesting is then that the Nikkor 80mm EL lens is a stop faster. As for taking photos restricting the aperture will have an effect on the sensor.

The Nikkor is f5.6 70% at 360nm vs the Soligor is f3.5 45% at 360nm. Dropping the Soligor to match for a photo will restrict the light.

 

Jonathan is the scale for the previous test comparable to the last test?

You have the UAT at 68% at f4.5 vs the Nikkor 80mm EL at 70% at f5.6 at 360nm. That places the Nikkor roughly 2/3rds of a stop better. Is that correct? That is basically what I saw at 370nm. But I am unclear about my true apertures.

[Andy Perrin]

Dabateman, the question it would answer is whether their equipment is working properly. The results SHOULD be independent of aperture but that doesn’t mean they are.

[ulf]

Dabateman, the question it would answer is whether their equipment is working properly. The results SHOULD be independent of aperture but that doesn’t mean they are.

That is correct. What we are trying to measure here is the transmission of the glass, totally unaffected by the aperture.

Therefore it is a good idea to perform the measurements with a fully open aperture to avoid the aperture interfering with the measurement.

 

A good routine is to verify that is true is to close the aperture and see that the signal to the spectrometer is not dropping, when closing it slightly.

Eventually it will when the aperture becomes smaller than the beam.

 

A wide beam can be a good idea as that decreases the measurement time and lower the noise, but the beam cannot be too wide that it is cut by the aperture.

This is a part of the adjustment you have to do before the actual measurement.

There is plenty of time doing that when the light source is stabilising.

[dabateman]

Ok I see the error in my reasoning.

The pure transmission of the Nikkor 80mm EL is 82% at 380nm and 70% at 360nm. About 77% at 370nm.

Whereas the UAT is a fairly flat 70% at 370nm.

Thus the difference between these lenses is insignificant in photographic terms. Less than a third of a stop. So at the same aperture, they should look the same at 370nm due purely to glass coating transmission.

The aperture would effect the lenses the same way. So no real difference.

 

[JMC]

David, yes, the graph in post #26 can be compared directly with post #21.

 

As Ulf mentioned, a lot of work goes into the getting the diameter of the beam right - not too small as to make data collection times too long, and not too big as to risk hitting the lens iris. The grpahs may look simple but there is a lot of work that goes into them.

[Timber]

I see you are in London (me too), I have a few lenses to test if you are interested :D :P

[Andrea B.]

catching up here....

 

That is such an odd little dip in the UAT. However, unless one were using very narrow-band filters, then I can't see how it matters much.

 

Jonathan, I'm wondering if we would see this dip in a transmission chart for calcium fluoride?

[JMC]

Andrea, a while ago I bought some CaF2 windows I found while surfing the web one night. I measured the transmission of those here; https://www.ultravioletphotography.com/content/index.php/topic/2569-buying-2nd-hand-caf2-a-cautionary-tale/page__view__findpost__p__19211

 

They all showed a dip at around 310nm, which where the one for the UAT is. I'm not sure what causes it, perhaps impurities, or perhaps there are different crystal structures with CaF2? Perhaps the UAT uses natural CaF2 which was subject to variations? But yes, that dip does occur with some CaF2.

 

As you saw, it's not going to be a problem though in normal use.

[JMC]

I recently ran a bunch of lenses for Timber, including enlarger lenses and some of his primes, and he is happy for me to share the data with the forum.

 

A question now for the admin really. This thread was originally planned to be one where I could update on any ongoing work, as it was in development. I now believe I am happy with the method, so would it make sense to start a new 'lens transmission in the UV' thread, just covering the results of any lenses I test?

[ulf]

It would be easier to find the topic later if there was a one topic per lens, with a well chosen name.

As an initiator of that topic you could add and edit the first post as you do more tests to keep the data collected.

 

I would like to do the same with filter filter measurements I do, but have not yet gotten any OK for that idea yet.

[Andrea B.]

Jonathan & Ulf, yes, I welcome and support your ideas about presenting the info in a better format. I've just PM-ed Ulf about this.

 

Thank you for all the lens charts, Jonathan. SO good to have these.

 

---

 

The Rayfact eventually takes a minor dip past 250 nm. From the Rayfact brochure.

 

 

---

 

I'm getting hung up on "calcium fluorite" versus "calcium fluoride". Which is it anyway?

 

---

[Andrea B.]

So, if we know what the elements are made of, can we "predict" the transmittance, more or less?

For example, the UV-Planar, which reaches 320 nm, has 5 elements:

• 2 calcium fluoride elements (99% at 350 nm)
  
• one lithium fluoride ele (99% at 350 nm)
  
• one fluor crown glass Schott FK-3 ele (99% @ 350 nm)
  
• one flint glass Schott F1 ele (90% at 350 nm).
  

[JMC]

Andrea, I believe it is referred to as either 'calcium fluoride' or 'fluorite'. Not 'calcium fluorite'.

 

That dip below 250nm is I think not a temporary dip, but where it starts to drop down towards zero.

 

Predicting transmission would also need info on number of interfaces, any adhesives, and any coatings.

[nfoto]

---

 

I'm getting hung up on "calcium fluorite" versus "calcium fluoride". Which is it anyway?

 

 

A quick goggle search yielded

 

Fluorite (also called fluorspar) is the mineral form of calcium fluoride

[Andrea B.]

A quick goggle search yielded...

 

Were those Google goggles? :cool: :cool:

[nfoto]

Don't poke fun for errors in writing. It is actually not very nice.

 

On a side note, I observe a lot of spelling and grammatical errors from our natively English-speaking members. As an administrator I have the means of stepping in and correcting those errors, which I never have attempted. One has to read with a will to understand.

[dabateman]

I tell people I only know broken English. Yes my spelling is horrible many times. Auto correct doesn't help either on a phone as I think I have said ultraviolet or ultraviolent many times. The second one goes back to solving math problems with swords.

[Andrea B.]

oh Birna, surely you know me better than that.

It was just wordplay. I had thought it would make you laugh!! I like to make you laugh!!

So I apologize for my unwanted wordplay!!

 

---

 

When I have time, I correct misspellings and minor grammar errors for everyone.

(The Editor's Hat is around here somewhere unless I packed it already.)

 

---

 

ADDED Later, Off Topic:

 

UVP is a resource. UVP is read around the world. So, it is useful and even a bit important to correct minor problems (only such things as misspellings or subject/verbs mismatched in number) for the sake of clarity. Especially in any posts involving UV safety (which must also be as factually correct as possible.) I accept and use any corrections and clarifications from any member who takes the time to offer them. (Especially for my own writing.) I want us to be as good a resource as we can be. None of us is perfect alone. Together, however, we can maintain and improve our posts and Stickies and tutorials.

 

Now, back to our regular programming......

[Andrea B.]

((Added an off topic remark to the preceding post to further explain the importance of maintaining UVP as a resource.))

[ulf]

When I have time, I correct misspellings and minor grammar errors for everyone.

(The Editor's Hat is around here somewhere unless I packed it already.)

 

As I'm a bit dyslectic and my mind is running faster than my writing, I do a lot of editing and rearranging when I write.

That introduce errors, both grammatical and sometimes logical.

 

I warmly welcome any corrections of the grammar, without any prior notice, as those errors make me embarrassed.

If something I write looks unclear or not logical, please ask, as I really want to be clear with what I write.

[JMC]

A while back, I ran a few lenses for Timber, just to see what they all look like for transmission in the UV. He's said he's fine with me posting them on here, so am doing so in case it is useful for everyone else.

 

There's a bit of a range of lenses, fro enlarger lenses to normal camera lenses, and at some point we hope to split these out into 1 lens per thread in a new section. But in the mean time, here they are together with data between 280nm and 420nm.

 

 

 

I think it is safe to say, ignore the flat portions of the graph where they level off at the short wavelength side (where they don't drop all the way to zero for some of the lenses), I think this is due to one of the many quirks of array spectrometers.

[ulf]

I think this is due to one of the many quirks of CCD spectrometers.

 

But your model of array spectrometer have a CMOS-detector not a CCD-detector.

I'm sure you ment Array-spectrometers, Jonathan.

 

They have their quirks to overcome, but also a very short measurement time compared to monochromator-based spectrometers.

[JMC]

Corrected it Ulf. Yes, they have their uses even with their quirks.

[Bill De Jager]

On a side note, I observe a lot of spelling and grammatical errors from our natively English-speaking members.

 

That's pretty routine for a major fraction of Americans online thanks to our educational system, or rather the 51 separate systems of variable quality. On the other hand, I've also been seeing a lot of errors online in the last few years which look to me like autocorrect or autocomplete errors, or voice-to-text errors. Things like diffuse being used instead of defuse, for instance.

 

That said, Google vs. goggle is a humorous typo. Next time I make a mistake like that (I catch most but a few slip through) I'll be happy to laugh along with whoever catches the error.