This FOT is made from 90 million optical fibers fused and stretched. It magnifies whatever it is on top of by 3x without lenses. Each fiber is like an independent pixel, stretched 3x. Notice the "URY" magnified from the MERCURY on the book cover.,
It’s best seen in this short video I took.
This is the same multi-channel approach Mojo uses for optical links in data centers.
Another photon fun moment with an angle-independent retroreflector. Apollo 11 left one on the moon, so a laser from earth would bounce back to whence it came.
It’s spooky how it tracks your pupil, or in this case, the iPhone camera lens: youtu.be/ZdcE1sV_S2o?si=BkqMMbblu5G5Mwi6
More background
very cool.
Dear Grok: what is the speed (optical efficiency) of a very high strand-count optical fiber taper (FOT)? Is it related to the length of the taper, or the ratio of entry to exit width or both?
Answer: Optical efficiency (often called "transmission," "throughput," or "light transmission") of a high strand-count optical fiber taper (FOT) typically ranges from about 50–85% for collimated/light input, depending on design, with practical values often 60–75% or lower for Lambertian sources.
High strand-count FOTs (fiber optic tapers or faceplates with thousands to millions of fibers) are used for image magnification/demagnification in applications like image intensifiers, sensors, or CCD coupling. "Very high strand-count" implies fine fibers (e.g., 5–10 µm or smaller at the small end) for high resolution.
Key Factors Affecting Efficiency
Efficiency is primarily related to the taper ratio (entry/exit width or diameter ratio, often called magnification or reduction ratio), not directly to the physical length of the taper (though length helps maintain a gradual taper for better performance).
Taper ratio (most important): Transmission decreases roughly with the square of the ratio (area scaling) due to etendue conservation and increased ray angles leading to losses. Higher ratios (e.g., 3:1 or more) reduce efficiency more, as rays exceed the effective numerical aperture (NA) and leak out or cause focal ratio degradation (FRD). For example, larger input-to-smaller-output ratios concentrate light but increase output NA and potential losses.
Packing fraction/fill factor: Hexagonal packing of fibers gives a theoretical maximum of ~90.7% core area, but real values are lower (~70–85%) due to cladding, interstitial gaps, and multi-fiber bundling. This sets an upper limit on efficiency independent of the taper.
NA and ray behavior: NA is inherited from the base material but varies along the taper (effective NA increases toward the narrow end). Light injected at the large end can exceed acceptance angles in the narrow section, causing losses. Extra-mural absorption (EMA) fibers reduce crosstalk but can absorb some light.
Other losses:
Fresnel reflections (~4–8% per face without AR coatings; high-index glass like n~1.8 worsens this).
Core absorption (minor, wavelength-dependent).
Internal reflections/scattering (worse with steeper tapers or imperfections).
FRD (focal ratio degradation), especially in shorter or imperfect tapers.
Typical reported values:
Collimated light: ~65–85% (higher end with good AR coatings and low ratio).
Lambertian/diffuse sources: Often ~55–65% or lower.
Examples from astronomy/spectroscopy (single or low-count tapers, but principles scale): 35–70% depending on input F/# and ratio.
Length matters indirectly: A longer, more gradual taper (e.g., tens of cm for high ratios) reduces abrupt angle changes and FRD, improving efficiency vs. very short tapers. However, it’s not a direct scaling factor like attenuation in straight fibers.
For very high strand-count imaging tapers, manufacturers (e.g., SCHOTT, Incom, or custom like SZPHOTON) quote specs based on ratio, and custom designs trade resolution/efficiency.
Efficiency is never 100% due to the inherent packing and optical losses. If you have specific ratio, NA, or application details, efficiency can be modeled more precisely.
Grok: compare this efficiency to modern Zeiss high performance cinema and still photography lenses:
Answer:
For the Zeiss Otus 85mm f/1.4 (and similar high-end Zeiss Otus lenses), the optical transmission efficiency is approximately 65–75%, with DxOMark measuring a T-stop of 1.7 for this specific lens. Cinema primes (e.g., Zeiss Supreme Primes, ARRI/Zeiss Master Primes) often achieve T-stops very close to their f-stops (e.g., T1.3–1.5 for f/1.4), implying 70–90%+ transmission thanks to advanced multi-coatings.
ps: there is a future where single photo efficiencies in pure optical compute is possible…but that is a ways off – although well worth pursing for efficiencies that would shrink current mega-data-centers back to normal sized rooms with power draws of a few thousand or even hundres watts (like the human noodle which runs on less than 100 watts at full steam). sorry oif the above was TMI…i’m super into optics/photonics…
Leave a Reply