|
|
![]() |
|
|
By Ed Gubbins Apr 17, 2007 3:57 PM
In a recent article about cost declines in fiber-to-the-premises deployment, I wrote about how pre-connectorized fiber drops (the fiber that meets the house) helped Verizon Communications bring down deployment costs. Pre-connectorized fiber comes in a variety of set lengths, with connectors already attached to each end. It saves carriers from having to splice fiber on their customers’ lawn, but the pre-set lengths require them to use more fiber than they need. To West Coast carrier SureWest Communications, which doesn’t use pre-connecterized fiber, it’s not a good trade-off, I assume because they don’t buy fiber in the same volume that Verizon does and therefore find the wasted material more expensive than the labor. (I also discussed this in a great Webcast with Michael Render of RVA Market Research, who has new fiber data out this week.) In response to that article, I got an e-mail from a service provider employee saying, “While connectorized drops have many benefits, [being] cheap is not one of them.” The bulk of that e-mail follows: “The economics of connectorized drop boil down to a trade-off between material and labor, and the material costs consist of more elements than just the excess fiber necessitated by cut lengths. In fact, the drop cable is only pennies per foot and is really the smallest factor in the cost of connectorized drop. The two factory-spliced connectors, the drop storage add-on at the [optical network terminal (ONT)], and the receptacles for the drop connectors that are added at the ONT and the serving terminal add significant cost. In simple terms, the labor savings is one fusion splice, since each fiber terminating in the receptacle at the network end of the drop must be spliced to a port in the receptacle. “Just for the sake of comparison, let's say that drop is $0.20 per foot. Seventy-five feet of drop would run $15. For the classic buried fiber drop, you'd also need to add a splice case to separate the drop fibers from the rest of the distribution--maybe $35. Total material cost for the whole thing: $50. Since cut lengths of connectorized drop are sold in 25-foot increments, let's just say that I only need 60 feet. That's wasting a whopping$3 compared to the 75-foot connectorized drop. But that is just the tip of the iceberg. “For connectorized drops, you need the drop plus two factory-spliced connectors that add around $120, or a total of $135 just for the drop. At the ONT, you need a storage module for the excess drop ($21), and a receptacle to plug the customer end of the drop into, say $36. That makes the cost of connecting one customer$192. But wait! Those are just the variable costs. You'll also have to splice a receptacle into the network access point. The network unit is sized to accommodate the ultimate number of units the access point is designed to serve: 2, 4, 8, on up to 12 ports. That little beauty is slightly more per port than the single port at the ONT ($45 times 4, maybe, for a four-port unit), and it becomes part of the distribution network the day it is spliced. If you're looking at a 60% take rate, that means you're paying for at least one port that you will probably never use. $47 or $237? For that to make sense in terms of the way we do copper drop, you'd have to be looking at a $187 splice. Even at half the connectorized cost, your splice would have to cost $72 for the trade-off to work. “Unfortunately, the reality for some of us is more complicated. The problem is that, for copper drops, the service technician who works the order at the customer premises terminates or splices the drop. Mostly, that does not happen with fiber, and not because fiber-splicing is impossibly difficult. New technology has made fiber-splicing so easy that an untrained staff weenie can do it. The problem is trying to impose a converged network on a distinctly unconverged population. For those of us with union contracts, the service tech shall never splice fiber. That job is reserved for the fiber splicer, a situation that actually makes connectorized drops begin look attractive. The cost of windshield time for the onesie-twosie fiber splices and the complexity that a third dispatch adds to the provisioning process makes connectorized drop look downright good. “There are other plusses and minuses, of course. Customer service [staff] love the ability to lay temporary drop on the ground, either to provide service while a cut drop is being replaced or, more frequently, to meet a customer due date when the drop crew misses its commitment. On the other hand, stocking and keeping track of the cut lengths of drop (which come one to a box) is challenging, and the old-timers who remember early fiber connectorization are waiting for the gel to crystallize and the connectivity to go away. I've been told that there is no gel in the new connectorized systems, but only time will tell.” E-mail me at ed.gubbins@penton.com. |
|
||||