Jump to content
UltravioletPhotography

Steinheil Munchen Auto-Quinaron 35mm F/2.8


msubees

Recommended Posts

I got one of this. It is not in the UV lens list, but I did a quick test and it shows nice UV. at least as good as EL Nikkor 80mm, or better.

 

very sharp. front filter thread: 49mm?

rear seems to be exakta mount but I can hand held it to fit my nikon extension tubes and no signicant light leak so I fit the 2" baader filter to the front (almost tight, but need hands there also).

 

I can see uv pattern at high iso (handheld) for dandelion and geranium.

 

I also bought a Hektor 1:28/250 mm today...a gigantic lens with no aperture control!

Link to comment
igoriginal

Can you supply some photos, please, of both - the lens, and also the resulting images?

 

Also, keep in mind that a lens doesn't have to transmit deeply into the UV-A, just to see UV absorption patterns.

 

In fact, I have many modern lenses which still show a trace of UV patterns. Even if they only transmit UV down to about 390nm or so.

Link to comment
will do, perhaps monday. need to drive down to OH today to give 2 talks tomorrow. this lens is going for $175 now at ebay. the Hektor went for 1125! I got two steals.
Link to comment

Igor,

 

I uploaded some shots of this lens to http://ww2.beetograp.../Steinheil-35mm

 

as well as a UV shot and a VIS shot. both at F5.6. The UV is not as sharp as I liked but it might be because my hand is always on the lens attaching it to the Nikon extension tube+Nikon to 4/3 adapter. I found this is the only way that I do not have significant light leak.

 

The lens has a switch that allows the aperture to be adjusted directly, or through a cable (so can be wide open and then stop down to set aperture once cable signal comes, perhaps the first generation of "Auto" lens).

Link to comment
Also, I notice the same UV dark lines similar to your recent flower -- perhaps an artifact because there is a structural fold, thus allowing no UV reflection there?
Link to comment

It is sometimes difficult to decide whether you are seeing a shadow or a UV-dark area. "-)

But making several photographs under different lighting angles usually clears that up.

 

Zach, I'm going to add your 'find' to the Lens List.

But first can you comment more about the source of the light leak?

Perhaps covering the lens with a cloth of some kind would help?

 

I hope to see you add a flower to the database eventually !!

Link to comment

Andrea,

 

the "light leak" I was referring to, of course, was due to I do not have the proper connector and simply used my hand to hold the lens to the Nikon extension tubes. if one is not careful, then some light will come in. Initially i thought this lens had no UV transmission and later I realized I had too much visible light came in, if I only used the 4/3 to Nikon adapter and then added this lens.

 

I did not transmission test and it has some UV transmission from 400-350 but it falls significantly after that. The OD industries lens fared better than the Steinheil. I will add a graph here soon (need to enter all the data!).

 

I will try soon. I do have a decent photo of the Eastern skunk cabbage, so I might give the first species in Araceae! :) :)

I also noticed Easter lily is not in the database yet, which I have.

 

I am not good with plant identification to species, and most I have shot are common ones near my garden.

Link to comment

post-41-0-35040300-1402333293.jpg

 

here is my first transmission test ever :)

 

not as nice as kds since I did not try to get my chart to work and did not use SCAN mode. I manually changed wavelength and wrote down the data then used software to graph it.

 

Sorry, bad curve fitting also.

 

lens tested are: EL Nikkor 80 mm F/5.6 (old model with chrome base), 2 copies, OD Industries 50mm F/4.5, and Steinheil-Munchen Auto-Quinaron 35mm F/2.8.

 

Methods used:

1. Instrument: Perkin-Elmer Lambda 3B UV/VIS Spectrophotometer.

 

2. Machine warmed up for 20 min after powering on, and turning on vis and UV lamps.

 

3. I used a quarts curvette in the reference slot and placed the lens to be tested in the testing slot.

 

4. Most likely the transmission data is relative and NOT normalized and the quartz curvette might not have been used, since I normally use the "absorbance mode" which can be "zeroed" but I did not zero with the tranmission test.

 

5. I tried to obtain the highest transmission at 700 nm. This required me removing the tray for the Steinheil lens and adjusting the height of the lens so light would go through the lens, for the other three lens I only needed to add something (about 5 mm tall) underneath each lens.

 

6. Once I obtained this, I did not move the lens or open the machine again, but only changed wavelength manually to 600, 500, 400, 390, etc. and recorded the % transmission -- this means each lens is internally consistent (e.g. 85% at 700nm might only be relative but all transmissions at other wavelengths will be correctly proportional to this). Nearly all readings were stable at 0.1 % about 2 seconds after designated wavelength was obtained.

 

7. all lens are also directly comparable, since I did not use "auto-zero" between the tests. it is only that the 100%, I was not sure how that was defined since I never used "auto-zero".

 

One can see that the $10 OD Industries lens is not too bad. The new lens (my main test subject, was going to keep it performed at least as good as the ODI :) did not perform so well, but nonetheless, it will be a lens usable for UV photography. I wonder why ODI transmitted more light at 360-380 (>70-78%) than 400-700 (66-69%).

 

I am thinking of selling it at ebay even though the lens is a beautiful piece of art. Another copy sold for $224 today.

Link to comment

For your transmission test to be useful there should be a description of the methodology and equipment used. Ideally there should be an error estimate in a formal test.

But at least try for the first two suggestions. "-)

Link to comment

Andrea,

 

thanks. I added the instruments and methods there. I am not sure how to measure the error in this case. For biological samples, we do triplicates.

 

I may have to re do this after using "zero" button on the machine to obtain a 100% transmission somewhere.

 

For your transmission test to be useful there should be a description of the methodology and equipment used. Ideally there should be an error estimate in a formal test.

But at least try for the first two suggestions. "-)

Link to comment

I am a bit confused today after I tried to do another run, comparing X135 with ODI (off topic?). I seem to have different results depending on where I set the 100%. I tried to set nothing (air) to 100% first at 500nm. another method is to put the lens in, and then set 100% at the same wavelength.

 

ODI actually performed too good compared to X135. Yesterday's 69% at 700nm must be due to miss-alighnment (the lens has a smaller opening). i need to normalize in excel and see if the 2 curves align. Mathematically they should be? it should not matter where the 100% is, the shape of the curve should not change.

Link to comment
no worries. the normalized data from yesterday and today's remeasurement match quite well. graph presented to the original lens test thread (X135, ODI etc).
Link to comment
I have been working on this for a while. Calibration is the tricky part. Sounde like you might want to see if the beam underfills the lens pupil. I used to have a PE Lambda 6 with an integrating sphere, a sphere is required for lens quantitative transmission testing.
Link to comment

Yes, now I realized that the graph I just presented there might not be fair.

 

I think i need to

1). set transmission at a wavelength (say 400nm) to 100%, perhaps with air (or quartz)

2). fiddle with each lens to get the highest reading so i know the lens is not blocking some light path

3). do not touch lens and change wavelength or do a scan.

 

Does KDS give his methods somewhere?

 

I have been working on this for a while. Calibration is the tricky part. Sounde like you might want to see if the beam underfills the lens pupil. I used to have a PE Lambda 6 with an integrating sphere, a sphere is required for lens quantitative transmission testing.

Link to comment

At minimum run a transmittance calibration scan on air.

 

The caveat is that since your slit height is probably fixed you will likely clip the ends of it on all but the widest pupils, so all you can do is align the lens as best you can. Also, unless you have an integrating sphere the slit image may no longer match the detector geometry.

 

Unfortunatly, like my Lambda 6, your 3B incorporates a single grating scanning monochromator. One really needs a scanning double monochromator to measure across a steep UV cut off.

It is however incrementally better than the single fixed grating spectrometers used by others.

Link to comment

John,

 

I was curious if the 100% transmission stayed that way at different nm. it was close enough. I tried a few UV wavelength with air and it stayed within 5% of 100%.

 

I was almost ready to do for each wavelength: 1). set 100% to air, 2). measure with lens...3). change wavelength, repeat 1 and 2. but decided it is too much work (and finicky for some lens).

 

At minimum run a transmittance calibration scan on air.

 

The caveat is that since your slit height is probably fixed you will likely clip the ends of it on all but the widest pupils, so all you can do is align the lens as best you can. Also, unless you have an integrating sphere the slit image may no longer match the detector geometry.

 

Unfortunatly, like my Lambda 6, your 3B incorporates a single grating scanning monochromator. One really needs a scanning double monochromator to measure across a steep UV cut off.

It is however incrementally better than the single fixed grating spectrometers used by others.

Link to comment

Zach,

I occurs to me that you may not have computer control of the scanning on that old spec.

If you set 100% T at 400 nm and then manually record the reading at every 10 nm down to 300 you might have your correction curve. The transmittance then is the ratio at each point of course.

I also recall that on my old Perkin-Elmer spec it switches over from the tungsten lamp to a deuterium lamp around 320-340 nm.

Link to comment

John,

It does have a "Scan" mode and you can specify the scan speed, but not range. I tried it and it seems to scan down from whatever wavelength at the current setting (e.g. at 400nm, it will start at 400, and gets shorter). I did do a curve by hand and the error was about 5% max (at 300, 310, and 2.5% for the rest: 320 to 390). so I think my data is fine?

Link to comment

With no better instrument control or data acquisition, it may be the best you are going to get.

 

The bottom line is, you still only have a single monochromator and the measurement geometry will never really be right.

Link to comment

John,

 

At this point, I think the "integration sphere" might be more important though.

 

I read a few papers published in journals about measuring contact lens for UV transmission and they pretty much used the same method. Scanned a blank to show the transmission is near 100% then scanned lens....

Link to comment
Yes, the integrating sphere is a critical component of the measurement geometry. I used to have a Lambda 6 (not the one I own now) that had an internal sphere on the detector side. The throughput of a sphere is typically ~10% so it really costs you in terms of dynamic range since the lamps inside the spectrophotomer are not powerful enough to compensate. Before you go hunting for one though, I must reiterate that a sphere is not the only critical requirement. Like I said before the Lambda 3 is still just a single monochromator as well as other limitations.
Link to comment
It is quite expensive it seems for the sphere...i tried to look for it at ebay and elsewhere...minimum seems to be $1600. so I wont be able to afford one unless I get two more Leica lens cheaply and ebay them :lol:
Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...