Why doesn’t simple focal-length geometry predict magnification in macro photography?
Asked 1/24/2015
5 views
2 answers
0
I’m using a Canon 550D with an EF 100mm f/2.8 Macro USM. At the lens’s minimum focusing distance (about 310 mm from the sensor plane to the subject), the lens reaches 1:1 magnification, so roughly 22 mm of a ruler fills the 22.5 mm-wide APS-C sensor.
My confusion is this: if the lens is 100 mm focal length and the sensor-to-subject distance is 310 mm, I expected the geometry to work like a simple pinhole/thin-lens diagram, with about 210 mm from subject to the optical center and 100 mm from there to the sensor. That would suggest the image should be reduced by about 2.1:1, not reproduced at 1:1.
So where is the mistake? Why doesn’t the focal length divided by subject distance give the image size correctly in this macro case?
Originally by Photography Stack Exchange contributor. Source · Licensed CC BY-SA 4.0
Photography Stack Exchange contributor
11y ago
2 Answers
5
The focal length of a lens only applies to objects at infinity, and is the distance ( in the diagram below) between the lens and the sensor when such an object is in focus. The focal point (F) would actually be ON the sensor.
(image cropped from Wikipedia page)
As an object gets closer to the sensor, the focal point moves away from the lens, meaning the lens has to be moved away from the sensor to keep the object in focus. This is why many lenses lenses get longer as you focus on closer objects.
The distance from the sensor to the object in focus is called the focal distance, and is indicated by the scale on your lens.
The 1:1 ratio on your macro lens represents the size of the image on the sensor relative to the object's actual size, and since you have demonstrated you can see 22mm of the scale on the 22.5mm sensor, this is what should be expected (as 22:22.5 is more or less exactly 1:1).
It should be noted here that due to their construction, many lenses have a physical length which is longer or shorter than their focal length. Telephoto lenses are defined as lenses whose length is shorter than the focal length, using a combination of lens elements that mimic the effects of a single lens as in the diagram above. And extreme wide-angle lenses can have focal lengths of 20mm or less but where the rear element of the lens is more than 20mm from the sensor of the camera.
Originally by user456. Source · Licensed CC BY-SA 4.0
user456
11y ago
0
Generated from our catalog & community — verify before relying on it.
The mistake is treating the lens like a fixed thin lens with the sensor always one focal length behind it.
A 100 mm focal length only gives a 100 mm lens-to-sensor distance when focused at infinity. At close focus, the image plane must move farther from the lens, and macro lenses often change their effective optical geometry internally as they focus.
At 1:1 magnification, the image distance and object distance (measured from the lens’s principal plane, not the front of the lens or the sensor mark) are approximately equal. In the ideal thin-lens case, both are about 2f, so a 100 mm lens would be around 200 mm from the subject and 200 mm from the sensor at 1:1.
Also, the quoted 310 mm minimum focusing distance is usually measured from the sensor plane to the subject, not from the lens’s optical center. Since the lens itself takes up part of that distance, the actual object distance from the principal plane is much less than 310 mm.
So the simple ratio fails because:
- focal length is not the sensor distance at close focus,
- distances must be measured from the principal plane, and
- macro focusing changes the lens-to-sensor spacing and sometimes effective focal length.
Recommended products
UniqueBot
AI11y ago
Your Answer
Related Questions
Why doesn’t 1:1 magnification on a macro lens occur at 4× the focal length?
Does image size on the sensor depend on sensor size or 35mm-equivalent focal length?
Why doesn’t the simple focal-length formula match my measured subject size in a photo?
How do I calculate the focal length needed to frame a subject of a given size at a given distance?
How can I estimate the minimum magnification needed to resolve a feature of a given size?