If your rangefinder is not accurate how do you know you are focusing correctly? This is where the rubber hits the pavement, so to speak. All the focusing systems take their cue from what is indicated by the rangefinder system.
For this procedure you'll check the accuracy of your rangefinder against a known distance. The first one, infinity is easy, just take your camera outside or if you don't have too many trees in your neighborhood just point it out the window at a very distant object that has distinct vertical lines. Does the moving image line up with the fixed image in the rangefinder window? If it doesn't, make note of whether the moving image did not travel enough to line up with or it went too far past the fixed image. Keeping track of this detail will help later when we need to make a decision on the right corrective action.
For .9m or closest distance I set up a little apparatus that exactly positions the film plane .9m from a focus test target.
This focus test target was created by John Wilton and can be
downloaded from his web site at: http://www.ragarecords.com/photo/index.html.
Make sure you take pains to ensure everything is
perpendicular and true to the test target. As you see, you don't have to make an overly complex set up. I have a stick that has every distance position that is on the camera helical marked out along the sticks length. (.9m, 1.0m, 1.15m,... etc. out to 3m) Keep this stick around, it will come in handy for other tests.
When you are sure everything is set, dial in .9m and...
Have a look through the finder. What do you see? Again, does the moving image line up with the fixed image in the rangefinder window? Make note of how it far and which direction it is out of agreement. As we see with my example, this Kiev needs some work to its rangefinder as the moving image did not travel far enough to align with the fixed image.
If your rangefinder is out at both ends of the scale, it might make good sense to do another test at some intermediate distance like 3 or 4 metres to see if it confirms the other two findings. For the most part though, if Infinity and .9m are accurate then all the other values will fall in line.
With what we see here, we have some work to do, see
Adjusting the Rangefinder
November 11, 2008