f/# in reportEfficiency() = inf or NaN.
Closed this issue · 3 comments
1. Example with f/# = inf.
from raytracing import *
# defines the collection path, considering full image size and full NA.
path = ImagingPath()
path.append(olympus.XLUMPlanFLN20X().flipOrientation())
path.append(Space(d=180))
path.append(Lens(f=180, diameter=50.8, label="Tube lens"))
path.append(Space(d=180))
path.append(Aperture(diameter=18.83, label="Camera"))
print(path.entrancePupil())
path.displayWithObject(diameter=1.1, fanNumber=10, rayNumber=5, fanAngle=1, removeBlocked=False)
We think the f/# equals inf because the focal length of a telescope is inf. Raytracing automatically places the object at inf for a telescope (?), but I know that I want to place the object at the focal length of the first lens (here being the objective). Therefore, the f/# should be (focal length of objective)/(diameter of entrance pupil) = 9mm/4mm = 2.25.
2. Example with f/# = NaN.
from raytracing import *
path=ImagingPath()
path.append(Space(d=10))
path.append(Lens(f=10))
path.append(Space(d=10.01))
path.append(Aperture(diameter=2))
print(path.entrancePupil())
path.display()
Because the AS is placed exactly at the focal plane of the lens, the entrance pupil is at inf and has a diameter of in because raytracing puts the object af inf again. I know that the object must be at the object plane of the lens, so at the focal plane on the right side of the lens, so the f/# should be again (focal of lens)/(diameter of entrance pupil). Not sure how to find the diameter of the entrance pupil here though.
we could add a objectAtFocal=True in the report function, so that we just change the object position for the calculation ? Thoughts ?
The two systems shown above are telecentric systems: the stop aperture is at the focal point, so the entrance pupil is at infinity. In this situation, we have an indeterminate f-number.
According to Smith: the f-number and the numerical aperture are two quantities that define the same characteristic of a system. In general, one uses:
f-number: infinite conjugation systems ;
NA: finite conjugation systems.
In most telecentric systems, finite conjugation is used: the object is rarely placed at infinity. We will therefore use the following definition for the f-number:
f-number = 1/2NA, with NA = n'sinU', where n' is the refractive index at the image plane and U' is the angle of the illumination cone.
So I think that raytracing should not automatically put the object at inf?
f
Stale issue message