Dismantling the 35mm f/2.8 Ai

[John Geerts]

Yes, I made the 35/2.8 ai  and the 24/2.8 non-ai by swapping the aperture rings.

[richardHaw]

Yes, I made the 35/2.8 ai  and the 24/2.8 non-ai by swapping the aperture rings.

i see. if they have the same Ai kit then it should be OK
that was interesting by the way

[pluton]

Just a note about optical performance of the 6 element 35/2.8 K/Ai based ONLY on my two samples of the 6 element Ai version.
I recently took both lenses out for a comparative test spin and found that on a flat subject at about 3 meters, and again at about infinity, the area at the frame margins (the edge to about 5 or 6mm in) is noticeably blurrier than the center area. Stopping down improves contrast and therefore the impression of 'SDS' (small detail sharpness), but the slight blur...in comparison to center... never completely goes away.  I suspect, but presently have no way to determine, that in the later 5-element version, the designers may have sacrificed the distortion correction of the 6 element unit for improved edge SDS.  Subject for future research...

[Akira]

Just a note about optical performance of the 6 element 35/2.8 K/Ai based ONLY on my two samples of the 6 element Ai version.
I recently took both lenses out for a comparative test spin and found that on a flat subject at about 3 meters, and again at about infinity, the area at the frame margins (the edge to about 5 or 6mm in) is noticeably blurrier than the center area. Stopping down improves contrast and therefore the impression of 'SDS' (small detail sharpness), but the slight blur...in comparison to center... never completely goes away.  I suspect, but presently have no way to determine, that in the later 5-element version, the designers may have sacrificed the distortion correction of the 6 element unit for improved edge SDS.  Subject for future research...

Keith, thanks for sharing your real-world impression.  I've never used any Nikkor 35/2.8 from any generation.  So, I would hold off one until I receive any further info either from you or the others.

[richardHaw]

yaiks. OK, I will do a test when mine comes. I ordered 2 

[Hugh_3170]

Some of these possibilities are also alluded to in the Ai Conversion section of Roland Vinks excellent Nikon lens web site.

Handy to know information, since the factory made Ai rings are getting very hard to find now.

Yes, I made the 35/2.8 ai  and the 24/2.8 non-ai by swapping the aperture rings.

i see. if they have the same Ai kit then it should be OK
that was interesting by the way

[richardHaw]

Some of these possibilities are also alluded to in the Ai Conversion section of Roland Vinks excellent Nikon lens web site.

so, should i referencing it by the Ai kit number? 

[Roland Vink]

Reference for Nikon AI modifications:
http://photosynthesis.co.nz/nikon/aimod.html

Comment:
"The New Nikkors (K type) have very uniform styling but every model still has a different part, even lenses with the same general form and aperture range such as the 300/4.5 and 300/4.5 ED, and the 24/2.8, 28/2.8 and 35/2.8. It might turn out that some of these can be interchanged with little or no modification."

I haven't had a chance to check whether this is true or not...

[richardHaw]

I see. the older generation is very different. I have opened no less than 30 of them and sometimes, from the same model there would be small differences inside like how the objective was secured, if a part was chromed or not, tolerances and how the aperture fork worked,etc. very small details but enough to prevent people from interchanging parts.

[John Geerts]

Just a note about optical performance of the 6 element 35/2.8 K/Ai based ONLY on my two samples of the 6 element Ai version.
I recently took both lenses out for a comparative test spin and found that on a flat subject at about 3 meters, and again at about infinity, the area at the frame margins (the edge to about 5 or 6mm in) is noticeably blurrier than the center area. Stopping down improves contrast and therefore the impression of 'SDS' (small detail sharpness), but the slight blur...in comparison to center... never completely goes away.  I suspect, but presently have no way to determine, that in the later 5-element version, the designers may have sacrificed the distortion correction of the 6 element unit for improved edge SDS.  Subject for future research...

I am not sure what you exactly mean. I haven't noticed anything serious, but I didn't really test it thoroughly.

This is an uncorrected file from the 35/2.8  serial 791138

[pluton]

John, here's a full D800E resolution file that shows what I saw.  Shot at f/8 and long focus distance...the closer 3-meter test shots are deleted.  Both my lens samples performed identically.
At small size or low resolution output, it doesn't matter.  But if you were constructing compositions where it had to give the impression of constant "sharpness" all the way from the center to the FX frame edge, it might matter.

[richardHaw]

Now that I have both lenses. I can probably make a test as well.

I did check for the smearing but it seems to be OK on mine. I may have to do some more tests. Can you please give me some parameters to work with?

[pluton]

Find a convenient wall or side of building that has lots of texture.  Tripod about 4 or 5 meters away, camera sensor parallel to wall, live view focus (or your best eye focus) at the center only, shoot f/2.8, 4, 5.6, 8, etc. without changing the focus at all.
Repeat with second lens, without moving camera in any way.
Repeat, if possible, in a different location featuring long distance subject, say, buildings across a wide street or a rooftop view of the city.
Here's my convenient testing wall, shot with iPhone:


 By the way, thanks for sharing your lens dismantling experience here and in the blog!

[richardHaw]

Thanks, Pluton!

silly me, I was dismantling a different lens for this post! but nevermind, I will write a proper dismantling of the earlier 35mm f/2.8 as both lenses are very different mechanically.

[pluton]

At the risk of going overboard with lens comparisons after already have committed the infraction of going off subject, here is a link to a Dropbox folder of full size, minimum compressed JPEG files of a series I did this afternoon at my usual test spot.  D800E, Aperture Priority AE, ISO 100, tripod, exact same raw treatment to all. Featuring the 35/2.8 K/Ai, the 17-35/2.8@35mm, the Zeiss ZF35/2, and the last [and, the last choice] lens I have capable of 35mm, the Zoom-Nikkor 25-50/4@ 35mm.  The file names contain the necessary ID for the frames.
https://www.dropbox.com/sh/wmvl0w890kw5z77/AACra_JYhRzBaLlzLuQqqorWa?dl=0