Focus testing procedure (FF/BF).

Started May 5, 2012 | Discussions thread
Flat view
Stepanfo Senior Member • Posts: 1,168
Focus testing procedure (FF/BF).

Some people have been reporting recently some problems with autofocus, unfortunatelly failing to provide any meningful information. No, it is not enough to say “I have some problem with autofocus” or “My left points are much worse than the others.”

It is necessary to define, WHAT problem it is:
Is it front focus?
Back focus?
Is the focus more erroneous with some focus points?
Something else?
And what is the extent of this problem .
In case of FF/BF the extent is defined in numerical value of AF fine tune.

Proper back/front focus testing procedure.
Set camera to:
AF-S focusing (single frame, not continuous)
Single point (NOT 1 point with surrounding pinits, just 1 point alone!)
Set aperture to the widest setting.

Drive focus manually away from the focus distance to the end of the scale.
Switch AF back on, if it was necessary to switch it off when driving away.
Observe, where did it focus.
Set AF fine tune to some value.

Repeat as many times as needed to figure out correct AF fine tune value, focusing from both ends.

Write down the value of AF fine tune for the point you tested.
Repeat for all 11 points in 11 points selection.


If neccessary, test other points (for example from some problematic area).

The map can reveal for example AF module tilt, which can be repaired.

When publishing these values, provide not only what lens you used, but also at what distance was the target.

Target has to be so clear, so that all random AF fluctuations are if possible eliminated.
Target cannot contain any repeated pattern.

This is a factory Nikon target from their lens service manual:

This target is shot from an angle to be able to see, how I made it. While testing you have to photograph it in axis. I am writing this because already somebody did not get it.

Dimension of target chart:

height 70 mm,
strips from left to right 64, 3, 6, 9.5, 2.5, 6, 64 mm

Dimension of resolution charts (not important): 48x35mm

What exactly is on resolution charts does not matter, althought for assessing blurr, it is better to have white marks on black background.

I think it is quite obvious that the charts must be matt black, not printed on some glossy surface etc.

Distance of the chart to the sensor position mark on the camera body:

For DX cameras
Distance in cm = 4* focal length in mm.
18 mm focal lenght gives 72 cm distance, 200 mm gives 800 cm distance etc.

For FX cameras:
Distance in cm = 3* focal length in mm.
50 mm focal lenght gives 150 cm distance, 105 mm gives 315 cm distance etc.

In Nikon service manuals, from which the data is, the resolution charts are 45mm apart. This has connection only to the numbers put in the service software, so the distance of the resolution charts should be set the way you need it to be able to adjust focusing effectivelly.

The camera – target distance is of course not mandatory too, but gives you an idea in what conditions lenses are tested in Nikon repair facilities and you should use it if possible.

This is a chart used by Nikon service facilities. If you want to talk with Nikon about focus problems, you have to use the same target in principle, and if possible under the same conditions as they use it.

I spent some time creating this post and I think it is quite usefull.

I am not sure if it should not be posted in other forums sections too or if some article should not be made from it, so that it just does not get lost in other threads.

Flat view
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow