How do I choose an f-number when designing a camera system?
Asked 10/31/2022
132 views
2 answers
0
I’m designing a camera for a project and I know the basic relation f-number = focal length / aperture diameter. I’ve already estimated sensor pixel size and chosen a focal length based on the resolution I want on the target. How should I choose the lens f-number or aperture size? Is there an “optimal” value, or do I pick it based on field of view, light level, and available lenses?
Originally by Photography Stack Exchange contributor. Source · Licensed CC BY-SA 4.0
Photography Stack Exchange contributor
3y ago
2 Answers
1
Assuming you're designing the internals of the camera, it isn't completely unreasonable to simply specify a "reasonable" figure for focal length (to give the desired field of view) and aperture (to provide sufficient light) and call out a commercially available lens with those specs. For instance, if you had an old 1/3 inch (diagonal) 4:3 sensor and wanted a "normal" field of view (30-35 degrees corner to corner) you'd specify a lens in the range around 8-10 mm focal length and f/4 or so (maybe faster if you wish to design in low light capability), then look for commercially available lenses in with those parameters. Given that's a fairly common range for video cameras, you shouldn't have any problem finding a make/model to call out "or equivalent".
Now, if you want/need to design the actual lens, that's a much more complex problem and would require you to spend considerable time with optical design ray tracing software, modifying existing designs to meet your specific needs -- but in that case, your professor should have at least introduced you to the school's copy of such software (it typically costs thousands of dollars for a license, so you probably aren't expected to buy your own) and given a starter on how to use it.
Originally by user89902. Source · Licensed CC BY-SA 4.0
user89902
3y ago
0
Generated from our catalog & community — verify before relying on it.
There usually isn’t one single “optimal” f-number you calculate from pixel size alone. In camera design, focal length is typically chosen first to get the field of view you want, then aperture/f-number is chosen to provide enough light and the depth of field or low-light performance you need.
Using f-number = f / D, once you choose a focal length, the aperture diameter follows from the f-number you want.
A practical approach is often:
- choose focal length for the desired field of view
- choose an f-number that gives sufficient light gathering
- then select a commercially available lens with roughly those specs
For example, for a small sensor and a “normal” viewing angle, something around 8–10 mm at about f/4 could be a reasonable starting point, with a faster lens if low-light performance matters.
So instead of deriving f-number strictly from detector pixel size, pick lens parameters from system requirements: field of view, brightness, low-light needs, and available optics.
Recommended products
UniqueBot
AI3y ago
Your Answer
Related Questions
Why do cameras record f-number instead of the actual aperture diameter?
How can I estimate a lens objective’s numerical aperture from focal length and aperture?
Why does a smaller f-number mean more light, even if the aperture diameter stays the same?
Why is aperture area expressed using focal length and f-number instead of just πr²?
Why do some zoom lenses have a constant maximum aperture while others are variable-aperture?