Fun with Fiber

In theory, yes. In practice, I bet that the fiber cannot be all that long before various aberrations turn the image into a smear.

All the practical image-transmission systems using optical fibers use a lot of them in parallel, one per pixel.

Joe Gwinn

Endoscopes are pretty much all GRIN rod devices these days, I believe. GRIN rods don't suffer the same buildup of field curvature that cascades of relay lenses do.

Cheers

Phil Hobbs

The Edmund data sheet

claims a 25.4 degree "acceptance angle." If I plug in the refractive index of the core and cladding (wikipedia equation)

I get 12.8. This one

calculates to 24.7 full angle, with a largish error band.

This

sure looks like 25 degrees to me. I guess Edmund is using the full angle, while the rest of the world uses the off-axis half-angle.

Optics is fun, in moderation.

Run light pulses down the fiber with a radar-like setup. Now run the fiber thru various regions of temperate. Note with the proper instrumentation, you can "measure" the temperature over the length of the fiber. Real-time full length temperature measurements. Cool!

Yes; to monitor his data storage in the cloud..

That would be like transmitting an image underwater, or in a universe filled with gas; the cladding doesn't matter. But you couldn't bend the fiber.

Wouldn't work anyway, because at a fixed wavelength, all the spatial Fourier components of the image correspond to different transverse k vectors, and those in turn propagate at different angles to the chosen axis. No angular spread -> no image.

(There are subtleties involved, e.g. whether the Fourier components in view are those of the fields or of the image intensity, but both models predict an irreducibile angular spread for a given spatial frequency bandwidth.)

Modern GRIN rods are actually pretty good for some imaging applications, especially borescopes and endoscopes.

Cheers

Phil Hobbs

The NA is equal to sin(theta), where theta is the half-angle of acceptance. NA 0.22 is a standard value, corresponding to a cone of half-angle 180/pi*asin(0.22) = 12.7 degrees

Anything worth doing is worth doing to excess. --Meatloaf

Cheers

Phil Hobbs

Yes. But to be clear, these are not fiber-optic devices.

For more information, see US patents 3,257,902 and 4,168,882, both to Harold H. Hopkins of the UK.

Joe Gwinn

Well, the difference is mostly diameter.

Cheers

Phil Hobbs

And optical perfection. The second patent had chromatic correction, if I interpreted its drawings right - looked like two kinds of glass were used.

Joe Gwinn

Another interesting graded-index technology is Gradium glass lenses, where the index gradient is perpendicular to the face of the blank. You polish a convex spherical surface into the high-index face, and the index gradient weakens the lens at the edges, reducing the spherical aberration. (The edges cut further into the lower index material, so the refractive index at the surface is less.)

Being an optical systems guy and not a real lens designer, I've never specified any of those uber fancy tricks--lenses are something I buy and make do with.

Cheers

Phil Hobbs

Reconcile the Edmund data sheet and the Wikipedia equation and get back to us.

Interesting. I hadn't heard of this; don't know why.

A cheaper way to get the advantages of aspheric lenses, at least good enough for focusing laser beams and the like. The implication being that the correction isn't quite good enough to achieve diffraction-limited performance. I'll have to read deeper.

The fancy stuff only makes economic sense in cases where ultimate performance is required, never mind the cost, or the production volume is immense, so the required special tooling can be amortized of a huge volume.

For ultimate performance, see things like the Hubble (and it's spy-satellite siblings), or the optics used for mask making in semiconductor manufacturing.

For immense volume, see the little aspheric lenses used to focus the read laser in CD and DVD drives. Harold H. Hopkins invented this lens.

Otherwise, it's best to stick to the Edmund catalog.

Joe Gwinn

Nancy Drew? :)

Those are my wife's books, from when she was a kid. I was a Tom Swift fan, but I didn't keep any of the books.

--

John Larkin Highland Technology Inc

Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom timing and laser controllers Photonics and fiberoptic TTL data links VME analog, thermocouple, LVDT, synchro, tachometer Multichannel arbitrary waveform generators