Why is viewing the Sun through a telescope dangerous if a telescope can't increase surface brightness?
Asked 3/19/2024
2 views
2 answers
0
I understand that, in ideal optics, a telescope cannot increase an extended object's surface brightness; it only makes the image larger while keeping brightness per unit area roughly the same. But it is also true that looking at the Sun through an unfiltered telescope can quickly and permanently damage your eyes.
How are those two statements consistent?
For example, if a small lens projects a tiny image of the Sun onto paper, and that image is not hot enough to burn it, what happens if a telescope is added in front so the Sun's image becomes larger? Does any point on the paper become hotter, or is the same light just spread over a larger area? In other words, what optical principle explains why solar viewing through a telescope is so dangerous even though magnification alone does not increase surface brightness?
Originally by Photography Stack Exchange contributor. Source · Licensed CC BY-SA 4.0
Photography Stack Exchange contributor
2y ago
2 Answers
4
Assuming your eye is part of the optic system with the telescope, and assuming the exit pupil of the telescope's eyepiece isn't larger than your eye's iris (that is, your eye doesn't contribute as a secondary aperture stop), then a telescope+eye system is no different than any other imaging system. \$\require{cancel}\$
As it turns out, with some back-of-the envelope calculations, you can show that for any (lossless) optical system that produces a photographic image of the sun (that is, we're not talking about solar concentrators for solar thermal power), the intensity of the sun's image on the sensor is approximately $${12\over N^2}\,{\mathrm{W}\over\mathrm{mm^2}}$$
That is, for any square-millimeter section of the image sensor (or your eye's retina) that is completely covered by part of the sun's image (i.e., inside the sun's disk), that area is receiving thermal energy of about 12 watts divided by the f-number of the lens/telescope squared. That's a lot of thermal energy for your retina, and for most imaging sensors. Some cameras sensors can handle that intensity for a short period of time because the actual image area is so small (such as with smaller phone cameras, or for very wide-angle lenses that produce such a small sun image).
The maths of the above value are fairly simple. Rectilinear lenses (any lens used to image a solar eclipse will be rectilinear, because you're interested in apparent magnification) obey the pinhole projection formula, which is simply that rays entering the lens at an angle \$\alpha\$ to the optical axis exit at the same angle. This implies that a subject of diameter \$d_\mathrm{s}\$ in focus at a distance \$u\$ from the camera will result in a image of diameter \$d_\mathrm{i}\$ (with a rear focus distance of \$v\$) according to similar triangles: $$ {d_\mathrm{s} \over u} = {d_\mathrm{i} \over v} $$ For subjects very far away (such as the astronomical objects), the rear focus distance is essentially identical to the lens's focal length \$f\$: \$v \approx f\$. Thus, astronomical objects of diameter \$d_\mathrm{s}\$ at a distance \$u\$ from the Earth will result in a focused image of diameter \$d_\mathrm{i} = f\cdot (d_\mathrm{s}/u)\$.
The ratio of the Earth–Sun distance (\$1\,\mathrm{AU}\$) to the sun's diameter \$d_☉\$ is 107.5, or about \$108:1\$. So the diameter of the image of the sun on the camera's sensor is $$ d_\mathrm{i} = {f \over 108}\,. $$ Note that this is regardless of the lens's diameter, the lens's aperture size, or the size of the lens's entrance pupil.
For example, capturing the sun during an eclipse with a 500 mm lens will result in an image of \$500\,\mathrm{mm}/108 = 4.6\,\mathrm{mm}\$ on the sensor. Note that this is just the disc of the sun. This does not include the coronal halo, which is substantially larger, and usually of interest during an eclipse.
Roughly speaking, under direct sunlight on a clear day, the incoming solar radiation (insolation), the amount of power per area, is nominally \$\mathrm{I_s} = 1\,{\mathrm{kW/m^2}}\$. So aiming your camera directly at the sun, the lens's entrance pupil has a diameter \$D = f/N\$, and therefore experiences the power of \$\mathrm{I_s}\$ times the area of the entrance pupil: $$ \begin{align*} P &= \mathrm{I_s} \times \text{entrance pupil area} \\ &= \mathrm{I_s} \times \pi\;\left({D\over2}\right)^2 \\ &= \mathrm{I_s} \times {\pi\over4}\left({f\over N}\right)^2 \\ \end{align*} $$ All of that incoming power is focused onto the sun's image on the sensor. We already know the sun's image has a diameter of \$d_\mathrm{i}=f/108\$. So in terms of power per unit area of the sun's image on sensor, the sun-image sensor area experiences an irradiance of that power \$P\$ divided by the sun's image area (\$A_\mathrm{i} = \pi(\tfrac{d_\mathrm{i}}{2})^2\$): $$ \begin{align*} I &= {P \over \text{sun image area}} \\ &= P \times {1\over \pi} \left({2\over d_\mathrm{i}}\right)^2 \\ &= P \times {4\over\pi}\left({108\over f}\right)^2 \\ &= \mathrm{I_s}\times \bcancel{\pi\over 4}{\cancel{f^2}\over N^2}\bcancel{4\over\pi}{108^2\over \cancel{f^2}} \\ &= {1\,\mathrm{kW}\over \mathrm{m^2}} \cdot {108^2\over N^2} \\ &\approx {12\over N^2}\,{\mathrm{MW\over m^2}} = {12\over N^2}\,{\mathrm{W\over mm^2}} \end{align*} $$ This feels like a surprising fact, that regardless of the lens size, camera size, etc., the power intensity per unit area of the imaged sun on the sensor is only affected by the lens's aperture number. So a big DSLR with a fast \$f/2\$ lens results in \$3\,\mathrm{W/mm^2}\$, whereas an iPhone's \$f/1.78\$ lens results in 27% higher radiant power intensity, at \$3.8\,\mathrm{W/mm^2}\$. Of course, the difference is that the DSLR is producing that power intensity over a larger area on the sensor than the iPhone is. And assuming that the radiant heat distribution is approximately linear through the thin plane of the image sensor, the DSLR suffers from a larger Area/Circumference ratio of the sun's image on sensor. Thus, the thermal resistance to radiant heat dissipation issue scales, unfavorably, linearly with the lens's focal length.
Originally by user11924. Source · Licensed CC BY-SA 4.0
user11924
2y ago
0
Generated from our catalog & community — verify before relying on it.
They are consistent: magnification alone does not increase the Sun’s surface brightness on the image. A telescope mainly makes the solar image larger, spreading the same apparent brightness over a larger retinal area.
The danger comes from total power, not higher brightness per point. Compared with the naked eye, a telescope’s objective collects far more light because it has a much larger aperture than your eye’s pupil, then sends that light through the eyepiece into your eye. If the telescope’s exit pupil fits inside your pupil, your retina receives much more total solar energy than it would unaided, even though the image brightness per unit area is not higher than an ideal optical system allows.
So in your paper thought experiment: adding magnification by itself does not make any point hotter; it enlarges the image and lowers intensity unless aperture is increased correspondingly. But a telescope usually also has a much larger front aperture, so it gathers far more sunlight overall. That is why an unfiltered telescope is dangerous: it delivers a much larger total amount of concentrated solar energy onto your retina, damaging tissue quickly.
In short: surface brightness is limited, but total collected light scales with aperture, and that extra total energy is what causes injury.
Recommended products
UniqueBot
AI2y ago
Your Answer
Related Questions
Does sensor pixel size affect depth of field or the circle of confusion?
Why do lenses with the same f-number give the same exposure despite different sizes?
Can a full-frame lens be made effectively faster on a smaller sensor by shrinking its image circle?
Could a reflective layer behind 35mm film improve low-light sensitivity like an animal’s tapetum lucidum?
Why doesn’t a larger lens diameter change exposure if the focal length and f-number stay the same?