Lens Calibration

[Acorn]

I have just acquired a new Nikon 200-500mm lens that may require lens to camera (D7000) calibration. As is do quite a bit of bird photography and frequently shoot at the 500mm end, what advice regarding the calibration set-up, particularly the distance from object to lens, might forum members suggest. YouTube is not particularly helpful in this regard as videos have been produced by the usual crop of fast-speaking, gibberish individuals.

[Manfred M]

First of all, are you sure that it is a calibration issue?

I know a lot of bird photographers and wildlife photographers who use long lenses. I have asked them; only one I know of actually had a calibration issue with a lens. Focus errors (usually caused by using matrix focusing, rather than single point) and camera shake are the most common reasons I can think of that people feel that their lens needs calibration.

[Acorn]

To be honest, Manfred, I'm not sure. It might be a case of me familiarising myself with the new lens and perhaps issues as to whether VR should be switched on or off. In replacing my former Sigma 150-500 which I found somewhat soft either hand-held or tripod mounted, I have yet to use the new Nikon lens in conjunction with a tripod-mounted gimbal; this should give a more accurate assessment of the len's focusing accuracy. My initial use of the new lens was in conjunction with a bean bag and with VR switched on.

[Shadowman]

Barry,

If your camera has AF-Fine-Tune function then it is easy to make the adjustment, a bit more time consuming to actually verify the effectiveness. I viewed a few of those YouTube videos and they made it look simple but the button combo didn't seem to work for me, I think I may have the AEL button configured to do something else so nothing was being changed. You can manually change the +/-20 settings for up to 12 lenses but then you have to do the standard focusing test to see if the adjustment was warranted. As far as testing your own lens, start out with a stationary subject at your estimated distance you want to capture and start the test using a tripod, once you are comfortable that the lens can function with that method try something a bit more challenging such as a flag flapping in the wind; I'd then review the images on your monitor or do a few test prints.

[DanK]

If by calibration you mean micro-focus adjustments, this is independent of VR. It refers to looking for back- or front-focus. I would start simply by testing it. I haven't tested all of my lens-body combinations, but of those I have tested, only one needed any adjustment, and the adjustment was minor.

Any error is not likely to be the same at the short and long ends, so give your use, I would test at the long end. Canon recommends a minimum distance that I would have to look up (you can google this), but I think it may be 50 x FL. In your case, you are shooting at long distances, so I would test at a fairly long distance. Set the camera on a tripod, turn off VR, and set up a focus target on a diagonal. There are fancy ones available, but a yard/meter stick with fine gradations may be sufficient. Focus on one point manually, with live view magnified, and take a shot. Then do the same with AF, placing a single AF point on the point you used for MF. If the results are similar, relax and go enjoy your new lens.

In my one experience with a lens that could have used calibration, the difference was small enough that it really didn't matter in practice, but in some cases it does.

If you need to do it, one option is FoCal, which I used in the distant past. However, there are other methods that don't require a purchase.

[Manfred M]

To be honest, Manfred, I'm not sure. It might be a case of me familiarising myself with the new lens and perhaps issues as to whether VR should be switched on or off. In replacing my former Sigma 150-500 which I found somewhat soft either hand-held or tripod mounted, I have yet to use the new Nikon lens in conjunction with a tripod-mounted gimbal; this should give a more accurate assessment of the len's focusing accuracy. My initial use of the new lens was in conjunction with a bean bag and with VR switched on.

A new lens that is shipped by the factory should be operating "in spec", although mistakes can occur. I own the Sigma and it is about as sharp as my original version Nikkor 80 - 400mm "D" lens. I can attribute 100% of sharpness problems with my images to user error.

VR should be switched on when hand-held, but switched off when shooting with a steady tripod (assuming a stationary object); if you are tracking, say a bird in flight, I suggest that VR should be on. With a bean bag, I would test both ways as there could be a bit of residual settling unless that camera has been at rest.

Unless you are sure that there is a front or back focus issue and you know what you are doing (and have the appropriate equipment / tools for targeting), chances are you are going to make things worse, not better. As I said in my initial posting, test with a single focus point; that way you can be certain that the autofocus system is focusing where you want it and not on a tree branch or other object that the matrix focusing points have picked up. Select a suitable shutter speed and do your testing.

[DanK]

A new lens that is shipped by the factory should be operating "in spec", although mistakes can occur.

Yes, but my understanding is that in some cases, the manufacturing tolerances are such that if a lens that happens to be at one end of the distribution is Paired with a body that is at the other, the joint effect can make MFA worthwhile. However, I have only seen this described in vague, verbal terms. If one had data on the actual distributions, one could calculate the joint probability. My guess is that it is very small, and my experience, limited as it is, has been consistent with that.

Very low on my to-do list is to redo these tests, as I have purchased a few lenses since the last time I checked, some years ago.

[rpcrowe]

I have tested my Canon EF 100-400mm f/4.5-5.6L IS II lens and it is right on! I also found no problems with my new/old 28-70mm f/2.8 Tokina ATX Pro lens. I have been using my other go-to lenses, the 17-55mm f/2.8 IS and 70-200mm f/4L IS for so long that I have every confidence in them.

When I tested the 400mm f/5.6L lens (which I replaced with the 100-400 Mk.2), I realized that any focus problems with that lens were due to operator error.

[Manfred M]

Yes, but my understanding is that in some cases, the manufacturing tolerances are such that if a lens that happens to be at one end of the distribution is Paired with a body that is at the other, the joint effect can make MFA worthwhile. However, I have only seen this described in vague, verbal terms. If one had data on the actual distributions, one could calculate the joint probability. My guess is that it is very small, and my experience, limited as it is, has been consistent with that.

Very low on my to-do list is to redo these tests, as I have purchased a few lenses since the last time I checked, some years ago.

I can understand how someone who does not understand how specs are developed might come to that conclusion.

When specs are developed, the designers ensure that so long as the parts meet the manufacturing tolerances, the device will perform to spec. Specs are developed precisely for the reason you mention; the worst case negative tolerance part mated with the worst case positive tolerance part will still delivery in-spec performance. The most likely causes of performance failure will be assembly errors and calibration errors; both of which can occur, but would tend to be quite rare.

[royent]

Stick the lens on the camera and go and try it. Practice usually beats theory
Roy ( sorry if I sound a little blunt )

[Acorn]

Most of the replies to my topic suggest that I need to persevere and first test with a tripod and without VR. I hope it is just me and I don't need to go through a calibration routine. Many thanks to all who have responded, your helpful suggestions are much appreciated. If you see any of my images in the future I hope they prove that I've heeded your advice.

[NorthernFocus]

Most of the replies to my topic suggest that I need to persevere and first test with a tripod and without VR. I hope it is just me and I don't need to go through a calibration routine. Many thanks to all who have responded, your helpful suggestions are much appreciated. If you see any of my images in the future I hope they prove that I've heeded your advice.

Barry, there is a simple way to check for gross front/back focusing errors and it can be done with or without a tripod though preferably with. Find a flat spot of well textured ground like grass or gravel or better yet something with a pattern like bricks. Set a target on the ground preferably something flat like a book or magazine with high contrasting text, photos, etc. that provide a good focusing target. Angle it a bit to make it as perpendicular as possible with the line from your lens. You can also use a small, fuzzy stuffed animal but something flat is better. Either way the target needs to be large enough to make sure the focus point stays on it. At 500mm you should set up 40 ft away. You want to be at eye level so there is a good angle with the ground under the target. Use AF-S focus mode with a single focus point. At maximum aperture take a few shots of your target. If it looks out of focus then look at the ground. If the ground is sharp in front of the target the lens/camera is front focusing. And of course if the ground is sharp behind the target the lens/camera is back focusing. Read you manual how to correct it. I recommend you move 5 increments at a time. Less than that isn't really noticeable without a more accurate target. With this simple procedure you can figure out if you have a problem worth worrying about and can tweak it back within the ball park.

There has been some discussion about making sure that you use good technique. If you are having problems caused by camera shake nothing in the photo will be sharp.

Also if you are having issues with focus on targets that are moving reasonably fast either toward or away from the camera, the D7000 doesn't handle those situations well. That model does not have predictive focus which adjusts for that sort of motion.

[NorthernFocus]

In the context of CiC being a learning forum(and for anyone who cares), I'll provide additional commentary on the topic of lens/camera calibration. What follows is strictly about how to do it. I have no intention of entering into the debate of whether one should do it.

For anyone interested in fine tuning their camera/lens combinations (on camera models with the appropriate functionality) there are products on the market to do that. Probably the best known/most widely used in the LensAlign target system by Michael Tapes Design. MTD also sells his own proprietary calibration software package, FocusTune. It can be used in conjunction with LensAlign or with any other target. The other software that I'm aware of is FoCal which is fully automated and requires tethering to a computer during the calibration process.

LensAlign
There are two versions available depending on lens focal length to be tuned. I'm not going to describe it in detail. Anyone interested can go to the website and read all about it. I will say that LensAlign is a visual system that requires subjective evaluation of the results. The product does come with a user manual with setup instructions and guidance on how to shoot test images and evaluate the results. I have used the telephoto lens version of LensAlign for several years with good results. I purchased it in order to tune a lens after I had identified a front focusing problem via the simple procedure that I described in my previous post. Using the LensAlign is a tedious process and ultimately requires subjective evaluation. Therefore it is only as good as the testing technique and evaluative eye of the user.

FocusTune
Again a detailed description of the software can be obtained on the web site. I do own a copy of FocusTune which I use in conjunction with LensAlign. IMO the software needs some tweaking to be useful for the average person. It doesn't work very well unless the target is illuminated by a constant light source. Even outdoors in clear sunlight the changing angle of the sun and imperceptible changes in atmospheric conditions cause significant variations over the course of a few minutes that it takes to shoot the necessary images for the analysis. That causes a lot of scatter in the data at best and potentially a trend induced by lighting conditions rather than focus shift. I find the LensAlign to be a much more useful and accurate tool. The biggest benefit of the software is that it generates copies of the test photos with red markers along high contrast edges. Those images make it much easier to subjectively analyse the LensAlign images.

FoCal
Per the previous, those interested can visit the website for detailed description. I have not used FoCal myself. However based on what I know from the web site and from discussions with those who have used it, if I were starting all over this would be my tool of choice. It is a fully automated process but does require tethering to a laptop. Setup is similar to the other systems but the actual calibration is pretty quick which eliminates the variability from changing lighting conditions.

Hope this is useful to someone.

[AlwaysOnAuto]

A little off topic here, but I was under the impression, after reading about how it works, that because of the way VR works on a D7000, you didn't need to turn off VR when on a tripod. I have a D7000 and an 18-200 VR lens and have never turned it off when using my tripod without it being a problem.
Comments?

[NorthernFocus]

A little off topic here, but I was under the impression, after reading about how it works, that because of the way VR works on a D7000, you didn't need to turn off VR when on a tripod. I have a D7000 and an 18-200 VR lens and have never turned it off when using my tripod without it being a problem.
Comments?

On all Nikon DSLRs VR resides in the lens, not in the camera. Nothing special about the D7000. The VR in some lenses is tripod compatible. The 18-200 may be one of those. Check the manual.

[odds]

...18-200 VR lens and have never turned it off when using my tripod without it being a problem.

I am tempted to refer to two of Thom Hogan's articles. The first one is new, the second is eight years old (likely updated though).

First http://www.dslrbodies.com/newsviews/...-answered.html

There you will find Thom Hogan's comment on ".....I don't see any differences...":
Variations of this comment are common. Every time I see the words "I don't see..." I have to start asking questions, unfortunately. One is this: who taught you to see? Another is: what can you actually see?

Second http://www.bythom.com/nikon-vr.htm
That old article is probably still among the most informative on Nikon VR.