Talk:Gap loss

Breaks, dirt and outages
"Gap loss is usually caused by poorly fitted multi-sectional fiber optic cables, but can also be caused by breaks in the fiber optic line. If the line is directly buried, then dirt will usually fill in any gap, causing a signal outage; however, if in a sealed cable liner, a break may leave the two cable ends lined up enough for gap loss to occur.[3]"

I know this text is referenced with a footnote, but it is still incorrect: --A. B. (talk) 08:34, 10 January 2007 (UTC)
 * First, no dirt is going to get into a buried cable unless the cable is breached.
 * Any break in the fiber will 99.999% of the time result in a total outage even if the broken ends remain more or less aligned. This is true whether the cable is buried or aerial.
 * "sealed cable liner" is unusual wording. Is the writer trying to say "cable"? "buffer tube"? "tight buffer"?


 * I was going by what the reference books said about it. I've edited the article, but let me explain here:


 * 1) There is a difference between the cable and the fiberoptic line. when I say a break or breech of the line, I'm usually talking about PC lines (phsyically connected ones) which were joined poorly and have come apart.


 * 2) I was trying to say with the dirt that if the cable is buried directly, any break will probably ruin the whole line. If the cable is laid into a liner, a protective concrete sheath found in areas where the FO cable can't be bundled with other cables or otherwise protected, that breaking the sheath may not break the cable.


 * anyway, edited the text. If you want to rewrite, rewrite. It's not "my" article. :) -- Elar  a  girl  Talk 15:10, 10 January 2007 (UTC)


 * Your edit looks great. I didn't mean to sound like I was fussing. I wanted to raise the question on the talk page before just unilaterally changing stuff. (I should probably be more bold!)


 * I'm surprised that the reference said that. Having seen many, many broken fiber cables (mostly outside), I don't think I've ever seen a fiber break that didn't lead to a complete outage on that fiber.


 * I can see how in rare cases (<<1%), a tight-buffered indoor cable might do this).


 * I'll be curious to check out the book when I'm in the bookstore; for now it's not on my amazon list!


 * Thanks again for working on this article. We always need more stuff on practical fiber optics. --A. B. (talk) 15:41, 10 January 2007 (UTC)

There seems to be some confusion (here and in the article) about the meaning of terms like "break". There is a difference between breaking a fiber, and breaking an entire cable (like with a backhoe). Different again is a "break" resulting from polished fiber ends that should be butted together (e.g. in a mechanical splice), but which have a gap between them. In the latter case, one can have significant transmission across the "break".

I agree with A.B. that dirt is irrelevant. If the cable is so badly broken that it would be possible for dirt to enter, then it won't matter whether the cable is directly buried or is in some kind of duct or conduit. No light will get through. I can imagine, on the other hand, that if a fiber (core and cladding) broke inside an otherwise intact cable, the buffer and cable materials might hold the ends of the fiber together well enough to allow some transmission.--Srleffler 23:45, 10 January 2007 (UTC)

Hmmm, another thought: most of what I wrote above is imagining standard singlemode telecomm cables. Multimode fiber would be much more likely to exhibit non-total gap loss in the event of a breakage. --Srleffler 00:00, 11 January 2007 (UTC)

Enclosed gap
The article only addresses free-space unenclosed gaps. If a fiber optic connection is loose the gap will be enclosed by a cylinder. The same would be true of various splices. The enclosing material may have various optical reflecting/absorbing properties at various frequencies. The light would mostly hit the gap wall at just a slight angle -- does this make it reflect better? The bigger the gap, the more the enclosure would matter. It would help multimode more than singlemode. A U-shaped tube open at the top would allow easy fill of the gap with gel or liquid, observation of the gap size etc while still reducing the loss somewhat, except in cases where the gap is so large that the light would try to bounce at least several times within the enclosure.

What are typical fiber optic connector wall materials, what are the reflecting properties of these materials under typical loose-connection conditions?-96.237.15.180 (talk) 12:28, 30 August 2010 (UTC)

usually no gap loss at detector
"Gap loss is not usually significant at the optical detector, because the sensitive area of the detector is normally somewhat larger than the cross section of the fiber core. Unless the separation is substantial, all light emerging from the fiber, even though it diverges, will still strike the detector."

Is this usually true? When is it less true?-96.237.15.180 (talk) 14:53, 30 August 2010 (UTC)

more narrow definition
A more narrow definition:
 * gap loss: 1. The power loss that occurs when an optical signal is transferred from one fiber to another that is axially aligned with it, but longitudinally separated from it.

Synonym: longitudinal offset loss. -96.237.15.180 (talk) 14:55, 30 August 2010 (UTC)