Can You Bend Optical Fibres Too Much?

[xunxine]

Although optical fibres do not break easily, is it possible to bend the optical fibre too much that light that travels within the optical fibre will strike the wall at less than the critical angle and thus escape from the fibre through the wall?
Pls enlighten me. Thanks!

(I came across this question in a test paper but i don't think it is makes sense cos i think the engineers who designed the optical fibres surely would have considered this issue.)

 

[Redbelly98]

Yes, it is possible, some fraction (but not all) of the light can escape at a bend. And the effect is visible, you can see light leaving the fiber near a sharp bend.

 

[xunxine]

So it is true?
What is the possibility & frequency of this happening in reality? How does this affect the signals being transmitted? They are simply lost?

 

[elect_eng]

So it is true?
What is the possibility & frequency of this happening in reality? How does this affect the signals being transmitted? They are simply lost?

The possibility and frequency depends on the person handling and installing the fiber. The usual term for this is called "macrobending" which is large nearly constant curvature bending. Often splicing trays have the fibers bent in a circle. Typical silica glass fiber can be bent to less than 1mm radius without breaking, but all if the light will be lost if you do this. Every fiber has an acceptable bend radius limit which must be obeyed.

When the light escapes the core of the fiber, due to bending, it goes into the glass cladding and eventually hits the outer coating. This outer coating has refractive index greater than that of the cladding glass; hence, the light escapes into the coating with no internal reflection possible. The light is mostly absorbed in the coating, which is lossy to infrared light. This design structure tends to prevent escaped light from getting back into the guiding core, which would be a negative effect since the travel time delay difference can corrupt communications signals.

There is another effect called "microbending" which is a periodic or random series of very small bends or stress along the length of the fiber. This can create a scattering effect which is seen as loss of guided light.

 

[mgb_phys]

Yes they are lost, but since only part of the power is lost it will probably still be received
It's a way of spying on cables without cutting them.

Generally commercial fibres have a jacket/cover which stops them bending more than the critical radius

 

[xunxine]

I see the light now. Thanks much!

 

[Danger]

I don't know what the state-of-the-art is, but the first VR interactive gloves used 'nicked' fibre optics, wherein the light leakage indicated the flexation of the fingers and the measurement thereof was relayed to the computer as an indication of the wearer's intent. I'm sure that it's obsolete for that purpose by now, but I can think of a lot of situations in which the phenomenon would be useful.

 

[mgb_phys]

A similar technique is used in strain gauges, the stretching of the fibre changes the transmission properties. You can even use the light return time to measure where along the fibre the stretch is - so you can run a single fibre and measure multiple strains.

 

[Danger]

you can run a single fibre and measure multiple strains.

Now that is freaking impressive. As usual, my knowledge of technology is a couple of decades behind reality.

 

[mgb_phys]

Optical fibres are great for sensor networks, they are cheap, don't need remote power, don't create or suffer from electrical interference and don't have any spark/explosion risk.

There is a whole field in making coatings where the refractive index changes slightly when some chemical bonds to it, you only need a very very tiny change in 'n' to change the transmission characteristics. You coat some of this magic stuff on the fibre in place of the outer layer at regular intervals and you can sense anything from bacteria, air pollution, chemicals given off from over heated wiring insulation in ducts etc.

You can even put different sensor layers at different points on the same fibre.

 

[Danger]

Now you're just making me horny. Stop it! :grumpy:

 

[dingpud]

Now you're just making me horny.

Too funny.

 

[fizz_it]

The strain sensors are called Bragg grating sensors.

http://en.wikipedia.org/wiki/Fiber_Bragg_grating