Hacker News Reader

Comment by ZeroConcerns | Hacker News Reader

There's no link to the data sheet of the actual cable, but, yeah, looks like this should not have happened in such a short timeframe unless there's something really funny going on in that room, like ambient temperatures above 50 degC.

Another thing that should not have happened is installing the cable in loops in this way: any 'building' or 'underground' type cable needs to be of the exact length required at the demarcation point, fastened properly to prevent movement and terminated on a proper patch panel (can be a one-port box-type thingy for small setups), from where you use regular patch cords to connect your equipment.

(Loops are definitely allowed though, but that use case is mostly for aerial fiber to enable repair splices, and there are some very specific bend-radius and strain relief requirements, which, again should be spelled out in the cable data sheet)

View on HN

> any 'building' or 'underground' type cable needs to be of the exact length required at the demarcation point, fastened properly to prevent movement and terminated on a proper patch panel (can be a one-port box-type thingy for small setups)

How exact is exact? :-) I once had to reterminate some fiber that was cut and terminated to exact length, which means there was literally two centimeters from the wall to the connector. I literally had to squeeze the fiber splicer up against the wall to have a chance at splicing on new pigtails, but I had two mis-cuts and I was hosed. :-)

> Another thing that should not have happened is installing the cable in loops in this way: any 'building' or 'underground' type cable needs to be of the exact length required at the demarcation point...

This hasn't been my experience with fiber entrance cables terminated by ILECs, Spectrum, and Lumen. They typically leave a significant service loop bound to the cable ladder or backer board-- usually 15-20 feet.

Depends on the type of cable assembly. If it's fiber strands inside a soft-ish plastic jacket (and most of the cable is in fact in conduit), a service loop is fine, albeit a bit pointless for most repair scenarios. For armored cables (which are significantly stiffer), you only do these loops in situations where you expect to need to replace significant sections (think 'getting hit by a falling tree' or 'particularly aggressive rodents') and you have the space.

Both tree and rat took out my fiber so the loops are definitely useful. If your fiber goes through your whole house it's significantly less work to only have to reconnect one end instead of redoing the whole run.

I have a > 75ft service loop on a 48-count underground burial fiber from the street.

Thanks, I really appreciate the SMEs commenting here. I'm learning a lot.

Definitely learnt it the hard way this time. You're right that buried cables should be exact in length and fastened to a patch panel. I'll probably look into better conduit design as well for the next time (in 15 years?). Having shared conduits means I would risk damaging other cables if I tried to pull a new cable through.

Good conduit and patch panel design is definitely key for a happy life. Leaving some extra space/capacity initially is also a good idea, especially since (unless you're covering truly great distances) there's not exactly a lot of innovation in the single mode fibre space: strands you put in today (even if it's 'the cheapest stuff your vendor sells most of', which is generally my philosophy for selecting cables) will still be viable a few years down the road.

Sharing/in-place-repurposing conduit is not something I'd recommend, but if you must, leave a few dummy cables (a.k.a. 'pieces of string') on the initial install...

From one of the photos, the cable spec "G657A2" is visible on the outside - and specs listed for that indicate it's "bending insensitive single-mode fibre", apparently it can tolerate 10 loops around 15mm mandrel. (Which does surprise me).

But yes, agreed, a lot of "Er... why would you do it like that?" bits.

Those 10 loops definitely only apply to the single mode fibre itself, not the entire assembly with armor and everything, because that's just... physically impossible.

Cables for direct burial only like to be bent once or twice, and then only gently. Anything else may very well break the armor (whether plastic or metal), after which all bets are off.

Still, for the outer jacket to become brittle to the extent described, something else is required, which may very well turn out to be "shoddy manufacturing"...

>Another thing that should not have happened is installing the cable in loops in this way: any 'building' or 'underground' type cable needs to be of the exact length required at the demarcation point

This is awful advice I would discard immediately. It's poor practice and against code.

When pulling cable, especially fiber, the ends of the cable should be able to reach the fathest corner of the room. Excess cable should be in a service loop, properly secured to a wall, and terminated on a patch panel. Both ends of the cable should follow this rule. That means you're typically pulling cable that's 15m longer or more, depending on the room and configuration.

NEVER buy and pull cable that is the exact size. The cable literally comes from the factory looped up, it's designed to be looped (watch bend radius).

>Loops are definitely allowed though, but that use case is mostly for aerial fiber to enable repair splices

Again, awful advice that's against code. Underground fiber must have service loops at both ends, and must be terminated to a patch panel.

Reply on news.ycombinator.com