optical center vs geometrical center

Started 1 month ago | Questions thread
ForumParentFirstPrevious
Flat view
piticoto Regular Member • Posts: 343
optical center vs geometrical center

How much deviation can be between optical and geometrical (exterior) center? Anyone has some knowledge about this?

I just got 18-135 lens, put in on the old 5T, still waiting for A6100, and extended to 135 I took two pictures of a print about two feet away. I put the camera with the diameter of the lens resting on a heavy V block. The second picture was taken with the lens and camera rotated 180 degrees in the V block, basically the camera upside down. Identifying the center pixels of the two pictures, there is a horizontal shift of 138 pixels between the two immages at the center pixel, out of 4912 pixels immage width. Vertical was just few, what is seemed less than 5 pixels, I did not even count vertically. 69 pixels, half of 138 would be the deviation between optical and exterior diameter centers. I made the measurement twice, 138 is the average.

If let's say the sensor is not quite at the center of the E mount, 69 pixels means the sensor is off center by 0.33 mm. That seems a lot. It can be an angle deviation between optical center and geometrical center of the lens, not sensor sift to the mount. 0.33 mm over 135 mm focal length makes a small angle 0.14 degrees. Can this angle be from adjusting the lens, like front element, when it was manufactured to make sharpness even between opposite corners?

I know, in real life pictures I'm not going to see this. But then what if the corners don't have even sharpness? Take pictures of a chart to fill in the entire immage? How do I control perpendicularly of the lens to the chart, at home?

ANSWER:
This question has not been answered yet.
ForumParentFirstPrevious
Flat view
ForumParentFirstPrevious
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow