Calibrating focus on a Nikon FM
On my workbench, I have three broken Nikon FM's that I am cobbling together into one working unit.
In replacing the focus screen (which was scratched) I had to remove the metal cage that holds the screen,a few shims and a leaf spring to hold it all together.
But I didn't realize while taking it apart that there were little shim washers under the four screws that hold this cage to the body. The shim washers adjust the height of the cage that holds the focus screen, and therefore, affect focus calibration.
I think I got the right shims reinstalled, but I'm not sure. So I wanted to check the focus calibration.
Am I doing this right?
Camera on tripod, 135mm F2 mounted and pointing down 45 degrees towards a focus calibration chart. A bright desk light shines on the chart.
I taped a piece of plain ground glass (no fresnel) to the film plane on the back of the camera.
I then focused on the center of the chart, as accurately as possible.
Then, on bulb mode, open the shutter and look at the image on the ground glass with a loupe.
This showed the Nikon back-focusing about 12mm. I tried another camera, a Minolta SRT-102, and it showed a 10mm front focus using the same technique.
Should I be putting a shim between the ground glass and the film plane to account for the thickness of the film?
Is this method sound? Should I be adjusting for zero?
I'm not an expert in the field, but I would do is to use your method to establish a roughly exact positioning of the focusing screen, adjusting for zero error. Then I would repeat the test with film instead of ground glass, for fine-tuning. In order not to waste too much film in this fine-tuning phase, you might take a series of shots, with different focusing screen positions (noting down the screw turns, I suppose there are screws that help you in this kind of calibration).
After developing the film you find, again, the position with zero error among the 36 positions you have recorded (18 on each side of your previous zero point).
Just to make things easier, you would choose screw turns that fit nicely in this 18 tries per side. For instance, let's say that you know that a full 360° rotation of the calibrating screw/s will certainly be more than enough in one direction, you could do the following:
Zero point: picture 0;
20° clockwise: picture 1;
40° clockwise: picture 2;
60° clockwise: picture 3;
360° clockwise: picture 18;
Now you turn counterclockwise your calibrating screw/s, so as to be again in the "zero point" position.
And now you repeat the series in the other sense:
20° counterclockwise: picture 19;
40° counterclockwise: picture 20;
360° counterclockwise: picture 36;
This assumes that you extract 37 images from a 36 roll. Otherwise, you sacrifice one of the "one complete screw turn" settings, and you have exactly 36 images. I wouldn't sacrifice the "zero point" image anyway.
Hope to have written something that makes sense
I checked the focus accuracy of a FE, F, FT3 and EM by this method. I first taped a Type B screen on to the inner rails of FE body. Type B screen is basically a ground glass or really a plastic screen for FE and other upper models. I used a Nikkor 100/2.5 for the test at f/2.5 and focus the camera locking a cable release at B position and magnifying the ground glass image using a reverse 20/3.5 lens. Then I released the lock and checked the focus from the split image of FE. Split image was not in focus so I suspected that the thicknesses of B screen caused it. Second time I used a piece of totally flat fogged film taped on the inner rails of the camera and the camera was correctly focused.
Alternatively I checked the focus by positing a target at a given distance (preferably 4 ft) from the film plane. The film plane is indicated by a circle and a straight line passing through it at the left side of the speed dial of FE, probably similar on FM. This method also gives me an indication of focus accuracy of the body if the lens is in good condition. Lastly I checked the focus taking 5 or more successive bw photographs of a inclined target. This is the most dependable way and usually I never can focus 50/1.4 correctly wide open. Well good luck on this work. It is rewarding to restore old cameras.
An excellent idea, but the FM does not have adjustment screws for the height of the focus screen. It's got little brass shim washers underneath the screws that hold the assembly on the body.
Unfortunately, I can't figure out which shims were on there originally. They fell onto my workbench and mixed with parts from another body.
I guess the central premise of my issue is...
Does a ground glass on the film rails give you a different result than actual film?
Sponsored Ad. (Subscribers to APUG have the option to remove this ad.)
Originally Posted by Hops
I think there is no difference between a film or a ground glass as long as the ground side of the glass faces the lens.
Maybe - I dare say shamelessly - one can repeat the process using some material such as aluminium foil, the supermarket one, which should be of uniform thickness:
Each "sheet" should add a fraction of a millimetre. Very boring, but that should be only for "fine tuning".
You can "industrialise" the process by cutting all the little aluminium foil stripes first. Very nice if you have three of four hours that you don't know how to employ ;-)
Or you could ask information (on the internet, on this forum) about the thickness of the missing washers (measured with a calibre).
Anyway, to answer your later question, LF photographers do use this method, so I suppose it should be decently accurate if you aim for zero mistake, without taking into consideration the film thickness. The problem I see is that the glass is perfectly planar and the film might not sit perfectly planar on the focal plane, so the film method would take into account imperfections of your film pressor.
EDIT. Thinking better about it, planarity problems should not descend from the pressor, but from consumption, or imperfection, of the camera rails, which should be well reproduced by the glass method, if you press it against the camera rails. But if the rails have some shallow points inside (so as to make a slight "curve", as seen from above) and are correct at the extremities, the film might "copy" them (and make a curve) while the glass will not. In this case, you should see that the glass does not make an exact contact on all points of the rails.
So you could make like the dentists do: "paint" with some non-permanent marker the rails; put the glass on the rails; if the glass is "painted" along its entire lenght, the rails are fine and the glass method will work fine, aiming for zero error.
Last edited by Diapositivo; 01-05-2011 at 07:53 PM. Click to view previous post history.
Reason: Second thought
A very quick check - if the lens mount is in the right place wrt the body - is to focus on an object at infinity. If the lens focus ring is at infinity when the viewfinder image is in focus then all is OK.
And as a third thought, one might use the "dentist" trick (maybe with a white market) to check that the film pressor presses the film uniformly (painting the marker on the rails, closing the camera without film, opening the camera, I suppose one should see the white marker on the pressor for the entire rail lenght).
Thanks all for the suggestions. I think I will adjust it to zero on the ground glass, shoot a test roll and see what happens.
Another thing: I read somewhere that the angle adjustment on the mirror is not really the right way to calibrate focus. It makes some sense, as adjusting the angle would cause the light to travel different distances at the top and bottom of the frame. Are you supposed to adjust it for precisely 45 degrees?