AF Nikkor 80-200mm f/2.8 MFD back-focus workaround (of sorts)

Discussion in 'Nikon' started by chulster, Jun 10, 2020.

  1. Many of you are aware that the popular and venerable AF Nikkor 80-200mm f/2.8D lens has a back-focusing issue at and near the long end of the zoom range when asked to focus at or near the minimum focus distance. (It's well-documented.) I wanted to share a small discovery made yesterday regarding this issue.

    I was playing around with a 25mm extension tube attached to the lens, curious to see how close it could focus with this accessory. (For the curious, the answer is 41 inches or 104 cm from the sensor plane.)

    Everyone who's ever used an extension tube knows that the distance scale on a lens with an extension tube attached is completely useless as far as showing true focus distances is concerned. The indicated distance will be much, much greater than the actual distance.

    However, observing this once again gave me an idea. From previous testing, I knew that my copy of this lens, when set to 200mm, will autofocus sharply down to a distance of about 15 feet, but not much closer than that. The nominal minimum focus distance of the lens is 1.5m (about 5 feet).

    Now, I wondered, what would happen if—with the extension tube in place—I placed the camera at the native MFD from the subject? I did so. At this distance of 1.5 meters, the distance scale indicated a (false) distance of about four and a half meters, or 15 feet. Would the lens be able to autofocus accurately at 1.5 meters when it "thought" the distance was 4.5 meters, thanks to the extension tube?

    The answer was yes: the lens could autofocus accurately at the native MFD with the extension tube attached.

    Here are two photos for comparison. Both were taken using PDAF, at the MFD, but only the second photo was taken with an extension tube. Thus, the first photo reflects the normal AF performance of my copy of this lens at focal length 200mm at the MFD, while the second photo reflects how it performs at the same distance with extension tube attached. It's obvious that the focus is much more accurate (no back-focus) in the second photo. You may also notice that the extension tube had a salutary effect on pincushion distortion (no distortion correction was applied to either image). Click either photo to view it large.

    DSC_2189.jpg DSC_2194.jpg

    Before anyone takes umbrage, please note I am NOT advancing this as any sort of solution to the back-focus issue with this lens. It's just an entertaining and impractical workaround.

    By the way, here's one more photo, which shows the degree of magnification achievable by the lens with a 25mm extension tube attached. I had to focus this one manually, since AF is no more accurate at the effective MFD with extension tube than it is at the native MFD without one.

    Last edited: Jun 10, 2020
    Andrew Garrard likes this.
  2. Interesting!

    And, err, how big is your King? Normal playing card width?

    And is it at 200mm?
  3. Yes . . . It makes sense: The problem has nothing to do with the actual distance. The problem occurs when the internals of the lens are at their extreme. This should still occur, with the extension in place, if you move your subject so that the lens is again indicating at or near it's minimum focus distance.
  4. Yes and yes.
    mike_halliwell likes this.
  5. That is correct.
  6. Looks as if the lens has a front-focussing issue, not back-focussing.

    Presumably the playing card is leaning back against the books slightly, and in your first image the bottom of the card - nearer to the camera - is in much better focus than the top. Also, back focus would render the book spines in better focus than the card, which they're not. So the lens must be front focussing.

    Were those examples taken with the lens wide open?

    I'm trying to get my head round whether this is an optical issue - of the focus shifting when stopped down - or a mechanical/AF servo issue.

    Aberrations introduced by a shift of the IF group could easily cause a focus shift with stopping down.

    Whatever. It probably wouldn't occur in a mirrorless system using direct sensor focussing.
    Last edited: Jun 12, 2020
  7. A trick of perspective, perhaps. The camera was positioned above and to the right of the card. Thus, the closest part of the image is the top right corner, and the bottom of the card is farther than the top. The photo is back-focused, no question.

    The photos were all taken at f/4. I wouldn't characterize the issue as focus shift; this lens is quite free of that. The plane of subject focus does not move rearward or forward with increasing aperture.

    The close focusing error at or near 200mm is a quirk of this lens's focusing design, i would say. But not being an optical engineer, i can't speculate more specifically.
  8. Could be an AF servo error then.

    Just another symptom of Nikon's flakey DSLR autofocus.

    'AF Fine Tune' is not the answer Nikon. The whole system and its tolerances need to be re-thought.
  9. If only Nikon could come up with some kind of sitting that took the mirror and external AF system out of the equation... (I don't we'll see a fundamental dSLR redesign, unless the D6 already has one. My interest in improved AF fine tuning mostly concerns getting the best out of current hardware with affordable firmware upgrade - for the current and next dSLR generations - rather than long term, because I suspect Nikon are done with that.)
  10. They already did, just go and buy new camera and lenses:)
  11. I think we should tell big "thank you" to Nikon, for still supporting "venerable AF Nikkor 80-200mm f/2.8D lens".
  12. Yes. I very much like the ability to use lenses from the 70s (Ai) to the 20s (E). I like using old, cheap screw-drive lenses, warts and all.
    Nick D. likes this.
  13. I did, and it wasn't made by Nikon.

Share This Page