Outside plant fiber and FTTH construction field guide

What outside plant fiber construction is

Outside plant fiber construction is building the part of the fiber network that lives in the harsh outdoor world. Inside cabling runs in feet through a controlled building. OSP runs in miles through weather, water, traffic, and pole loading, from a central office or hub out to neighborhoods, and on to homes and businesses as fiber to the home (FTTH) or one of its variants (FTTx).

The work breaks into a handful of decisions made in order. You locate the existing utilities before any dig. You choose aerial on poles or underground in the ground. You place the cable along the route. You splice the fiber together at the joints and store slack for the future. You test the finished link with an OTDR and a power meter. Then you map the as-built so the next crew can find what you put in the ground.

Two governors run through every one of those steps and neither one exists inside a building. The first is the environment: water in the cable, ice on the strand, a car into the pole, frost heaving the conduit. The second is damage prevention. Every dig has to be located through 811 first, because the ground is full of other people's gas, power, and fiber, and a strike is somebody's outage or somebody's funeral. Hold those two ideas and the rest of OSP is craft.

The outdoor world and the locate ticket govern the job

Inside cabling lives in a clean, dry, conditioned space, and the work is about the cable. OSP work is about the cable plus everything trying to destroy it and everything already in the ground around it. Those two forces, the environment and damage prevention, drive more decisions than the fiber itself.

The environment is why OSP cable is water-blocked, armored, and rated for the temperature swing of a buried duct or an exposed strand. Water that gets into a cable wicks down the tube, and when it freezes it crushes fibers and spikes loss. Ice and wind load a messenger strand. Sun and frost cycle a pedestal. None of that is in play indoors, and all of it is in play the moment the cable leaves the building.

Damage prevention is the law half. You do not put a shovel, a plow, a bore head, or an anchor in the ground until the existing utilities are located and marked. This is not a courtesy. It is a statute in every state, enforced through the 811 one-call system, and the consequence of skipping it ranges from a fine to a gas explosion. Treat the locate ticket as the first line item on the job, not a box you check after the crew is mobilized.

OSP fiber vs inside plant cabling

The cleanest way to understand OSP is by contrast with inside plant. Inside plant is the structured cabling in the building: the riser, the horizontal runs to the work area, the cabling that the low-voltage and Class 2 rules govern. It runs in feet, in a controlled space, supported on a tray or in conduit, and the failure modes are bend radius, separation from power, and the jacket rating for the plenum or riser. The companion guide on low-voltage and Class 2 systems cabling covers that inside world.

OSP is the other side of the wall. It runs in miles, outdoors, buried or aerial, and the failure modes are water, dig-ins, pole loading, and the locate nobody called. The standards shift too. Inside, the NEC and TIA structured-cabling standards control. Outside on poles, the National Electrical Safety Code (NESC) controls clearances and loading, and 811 damage-prevention law controls the dig.

The cable itself is built for a different fight. Inside cable is light, riser- or plenum-rated, and dry. OSP cable is water-blocked, often armored, and rated for outdoor temperature and burial. The transition between the two worlds is its own detail. Most jurisdictions limit how far unrated OSP cable can run inside a building before it has to be spliced or terminated to a listed indoor-rated cable, so the entrance point is where inside and outside plant meet and where the rating rules change.

Why call 811 before you dig?

You call 811 because hitting an unmarked utility is the fastest way to kill someone or take out a neighborhood, and because locating first is the law in every state. The 811 one-call center notifies every member utility with facilities in your dig area, and each one marks its lines or confirms it has none. This is the single most important rule in OSP, and it is not negotiable.

The mechanics are simple and the timing is fixed by statute. You file a locate ticket describing the dig area, then you wait the legal notice period, commonly two to three business days depending on the state, for the utilities to respond. Many states run a positive-response system where each utility posts its status, and you confirm that every member has cleared or marked before a tool goes in the ground. The marks follow a color code: red for power, yellow for gas and oil, orange for communications, blue for water, green for sewer.

The marks are guidance, not gospel. A locate is generally accurate to a tolerance zone around the mark, often a couple of feet on each side, so within that zone you hand-dig or use vacuum excavation to expose the line before you machine-dig near it. The marks also expire, so a ticket that has aged out gets refreshed before work resumes. Hit a gas line because the ticket lapsed and the paint faded, and the fact that you called once does not save you. Verify against your state law and the local 811 center, because the notice period and the positive-response rules vary by jurisdiction.

Should the route go aerial or underground?

Route choice is the first big decision after the locate, and it comes down to what is already there, what it costs, and what the permits allow. Aerial means lashing the fiber to a messenger strand on existing poles. Underground means putting it in the ground, direct-buried or in conduit. Aerial is usually faster and cheaper to place where a pole line already exists, but it inherits the pole world: make-ready, NESC clearances, joint-use agreements, and exposure to wind, ice, and vehicle strikes. Underground costs more per foot and takes longer, but it is out of the weather and out of sight.

Most real builds are a mix. The route follows the poles where poles exist and the joint-use space is available, drops underground at road crossings and where there is no aerial path, and bores under obstacles. The engineer sets the route on the design, but the field confirms it, because a pole that looked attachable in the GIS can be full, leaning, or scheduled for replacement.

Cost is route-specific, so treat any rule of thumb as a starting point and price the actual path. Verify the route choice against the pole owner's make-ready findings, the permit conditions, and the engineer's design before you commit a crew.

The aerial build: strand, lashing, and attachment

An aerial fiber run rides a messenger strand, and the strand does the structural work. You attach a galvanized steel messenger to each pole, tension it, and then lash the fiber cable to it with a continuous stainless lashing wire run by a lashing machine that travels the span. The fiber carries no load. The strand carries the wind, the ice, and the weight, which is why the strand sizing and tension are engineered, not eyeballed.

Attachment hardware is specific and the pole owner specifies it. The strand lands on a bolt or a through-bolt with the correct clamp at the height assigned in the communication space. At corners and dead-ends the strand pull has to be balanced with a down-guy and anchor, because an unguyed corner pulls the pole out of plumb over time. Self-supporting cables exist, all-dielectric self-supporting (ADSS) and figure-8 with an integral messenger, and they change the hardware but not the loading question.

Slack is part of the aerial design too. You leave storage loops on snowshoe brackets at splice points and at intervals along the run, so a future splice or repair has fiber to work with without a re-pull. Verify the strand size, tension, guying, and attachment height against the pole owner's construction standards and the engineered loading, because the NESC and the owner control the pole, not the crew.

Pole make-ready and joint use

Make-ready is the work that has to happen on a pole before you are allowed to attach, and it is the part of an aerial build that wrecks schedules. A pole is joint-use property: the power company owns the top, the existing communications attachers own space below, and you are the new attacher asking for room. Before you hang a strand, the pole has to be surveyed, the existing attachments may have to be rearranged to open compliant space, and in some cases the pole is too short or too loaded and has to be replaced.

The process runs through the pole owner. You apply for attachment, the owner or a contractor performs an engineering survey on each pole, and the make-ready work and cost get assigned. Existing attachers may have to move their lines down to maintain separation, which means coordinating other companies' crews on a timeline you do not control. A single pole change-out can hold up a span for weeks.

The schedule lesson is to file make-ready early and treat the make-ready timeline as the long pole in the project, not the construction. The cost and the sequence are set by the pole owner's tariff and joint-use agreement, so confirm both before you promise a date. Federal one-touch make-ready rules speed parts of this in some jurisdictions, but the owner's process still controls.

NESC clearances: keeping out of the power space

The pole is divided into spaces, and the division exists to keep communications workers away from energized power. Power supply conductors ride at the top in the supply space. Communications, including fiber, ride at the bottom in the communication space. Between them is a separation that the National Electrical Safety Code sets to protect the worker, commonly a 40 inch safety zone between the lowest power conductor and the highest communication attachment. This is a safety clearance, and it is not where you cut corners to fit one more cable.

The clearances do not stop at the pole. The NESC also sets minimum heights for the cable over the ground, and they change by what is below: more clearance over a road or a railroad than over a pedestrian path, because a truck or a train needs room. Span sag from heat and ice loading has to be figured so the lowest point of the cable on a hot, loaded day still meets the ground clearance.

Get the vertical separation or the ground clearance wrong and you have created a hazard that an inspector or the pole owner will catch, or that a lineman will find the hard way. Confirm every clearance against the adopted edition of the NESC, the pole owner's standards, and the engineered design, because the exact dimensions depend on voltage, the attachment, and what the cable crosses. Hedge to the NESC and the owner here, always.

The underground build: direct-bury vs conduit

Underground placement comes in two forms. Direct-bury puts an armored cable straight into the ground, plowed in with a cable plow or laid in an open trench and backfilled. Conduit puts a duct or innerduct in the ground first, and the fiber is pulled or jetted through it later. Direct-bury is cheaper and faster up front. Conduit costs more but lets you replace or add cable later without digging the route again, which is why long-term routes and anything under pavement usually go in conduit.

Depth is set by code and by what is overhead. Common practice places fiber around 36 inches deep, with the NESC and local rules calling for more under roads and railroads and allowing less in some general areas. A detectable warning tape goes in the trench roughly 12 inches above the cable so the next excavator gets a warning before the shovel reaches the cable, and a tracer wire or a detectable element lets a future locate find a non-metallic cable.

The trench is only half the job. Restoration of the surface, the pavement, the curb, the landscaping, is a real cost and a permit condition, and a sloppy restoration is what the public and the city remember. Verify burial depth, warning tape, tracer, and restoration against the permit, the NESC, and local standards, because the required cover depends on the location and the jurisdiction.

HDD, boring, and the cross-bore danger

When you cannot open-cut, you go trenchless. Horizontal directional drilling (HDD) steers a bore head underground along a planned path, under a road, a driveway, or a creek, then pulls the conduit back through the bore. A smaller version, microtrenching, cuts a narrow slot a couple of inches wide and a foot or two deep in the pavement and lays a small duct in it, fast and with minimal restoration, where the jurisdiction allows it.

Trenchless work multiplies the locate risk, because you are drilling blind through ground that holds other utilities. The danger that gets people killed is the cross-bore: the bore path passes through an existing utility you did not see, most dangerously a sewer lateral. A bore through a sewer lateral can sit silent for years until a plumber clears a blockage, cuts into the conduit, and if a gas line shares the same mistake somewhere in the system, the result is an explosion. A common practice is to treat sewer laterals as high-consequence and to physically expose, camera, or electronically confirm separation, with a clearance such as 24 inches, before drilling near one.

The protection is the locate done right plus verification at the bore. Pothole or vacuum-excavate to expose crossing utilities, track the bore head, and confirm the path against the marks as you drill. Verify the cross-bore protocol against your state law, the 811 center, and the project specification, because the consequences here are not measured in dollars.

The OSP cable: loose-tube, ribbon, armored

OSP cable is built to survive the ground and the strand, and it comes in two main constructions. Loose-tube cable houses the fibers in gel-filled or dry buffer tubes that float free, so the fibers are isolated from the cable's stretch and from temperature swing, which suits long outdoor runs. Ribbon cable bonds fibers into flat ribbons, packing very high counts into a small cable and letting a crew mass-fusion-splice a whole ribbon at once, which is why high-count metro and feeder cables often run ribbon. The companion guide on data center fiber cabling types covers single-mode versus multimode and the OS and OM grades; OSP runs are almost always single-mode (OS2) for the distance.

Fiber count is set by the design: how many homes, how many splits, how much slack for growth. The reflex on a greenfield build is to place more count than today's demand, because pulling a second cable later costs far more than the glass did up front.

Outdoor construction adds armor and water protection. A corrugated steel armor layer resists rodents and gives the crush strength for direct burial. The jacket is rated for sunlight and burial. Where steel is a problem, near power or for lightning, all-dielectric cable does the same job without metal. The point of all of it is that an OSP cable is engineered for the environment first and the fiber count second.

Water-blocked cable and why water kills a run

Water is the quiet killer of outdoor fiber. Get water into a cable, through a nicked jacket or an unsealed closure, and it wicks along the buffer tubes by capillary action down the length of the cable. In a buried or aerial cable that water freezes in winter, and the ice crushes and microbends the fibers, which shows up as rising loss and eventually a dead link far from where the water got in.

That is why OSP cable is water-blocked. Older and many current designs flood the tubes with a water-blocking gel. Newer dry designs use water-swellable yarns and tapes that block water without the mess of gel, which speeds up splice prep because there is no flooding compound to clean off the fibers. Either way the cable is sealed end to end.

The cable being water-blocked only helps if the closures and the sheath stay sealed. Most water intrusion happens at a poorly closed splice case or a damaged jacket, not through the cable itself, so the field discipline is to seal every closure to spec and never leave an open end exposed overnight.

What is FTTH and how does PON deliver it?

Fiber to the home (FTTH) brings fiber all the way to the residence, and most FTTH is built on a passive optical network (PON). Passive is the key word: between the head end and the home there are no powered electronics, only glass and optical splitters, which is what makes the network cheap to operate over miles of plant.

The architecture runs in stages. At the head end sits the optical line terminal (OLT). From the OLT a feeder fiber runs out to the neighborhood. At a splitter, housed in a cabinet, a fiber distribution hub, or a closure, the single feeder signal is split passively to many homes, commonly 1 to 32 or 1 to 64 on GPON, and up to 1 to 128 on newer XGS-PON. Distribution fiber carries each split out toward the homes, a drop runs from the distribution to the house, and an optical network terminal (ONT) at the home converts light to the customer's services. FTTx is the family of variants: fiber to the curb, the node, or the building, where the fiber stops short of the home and another medium finishes the run.

The split ratio is a budget decision, not a free dial. Each split divides the light, so a higher split reaches fewer kilometers. A typical GPON loss budget around 28 dB supports a 1 to 32 split out to roughly 20 km, with higher splits trading reach for density. The engineer sets the split plan and the placement of the splitters; the field places and splices to it.

The fusion splice and the closure

OSP fiber is joined by fusion splicing, and the quality of the splice is the quality of the network. You strip the coating, cleave each fiber to a clean square end with a precision cleaver, align the two cores in a fusion splicer, and fire an electric arc that melts the glass into one continuous fiber. A good single-mode fusion splice loses very little light, commonly a fraction of a tenth of a decibel, and the splicer estimates the loss but the OTDR confirms it. The cleave is where splices live or die. A bad cleave gives a high-loss or reflective splice no arc can fix, so the cleaver blade and the prep are not where you save time.

The splice is then protected and housed. Each fusion joint gets a heat-shrink protector over a steel strength member, and the protected splices are organized in a splice tray. The trays live in a splice closure, a sealed enclosure rated for the environment, buried in a vault or handhole or hung on the strand aerially. The closure has to seal against water and stay re-enterable, because someone will open it again to add or repair fibers.

Slack is non-negotiable at every splice. You store service loops of cable on both sides of a closure so a future splice has fiber to pull into the tent or the truck without cutting the run short. Skip the slack and the next repair means a new section of cable and two more splices. Verify splice-loss acceptance, closure ratings, and slack lengths against the project specification and the TIA testing standards.

Closures, access points, and slack storage

The network has to be reachable, so OSP design builds in access points where crews can get to the fiber later. Underground, that means handholes and vaults: boxes set flush in the ground with a lid, sized so a splicer can work and a closure and its slack can be stored. Aerial, the access is the strand itself plus snowshoe brackets that hold storage loops. At the customer side and at distribution points, pedestals stick up out of the ground as the access enclosure for drops and small splits.

Where you put access points is a real design choice, because every splice point, every slack store, and every closure is a place the network can be worked on and also a place it can fail if it floods or fills with debris. Handholes get placed at splice locations, at major route changes, and at intervals long enough that a future pull or repair is reachable from one box to the next.

Slack management ties it together. The slack at each access point is what lets a future crew bring the fiber up to a splice trailer without re-pulling the span, and it is what gives a damaged section enough length to repair in place. Store it neatly on the racks and brackets the closure provides, log the slack length in the as-built, and the next crew can plan a repair from the records instead of from a shovel.

The drop to the home and the ONT

The drop is the last segment, from the distribution point to the house, and it is where the network meets the customer. Drops run aerial from a pole or pedestal to the home, or buried from a handhole or pedestal across the yard. Aerial drops use a self-supporting drop cable; buried drops use a toned, rugged drop, often placed in a small bore or a shallow trench across the lawn with the homeowner's surface to restore.

At the house the drop lands at a network interface device (NID) or an optical demarcation, and the fiber runs to the optical network terminal (ONT). The ONT is the powered box that converts the light to the customer's internet, voice, and video, and it is the only active device on the customer end of a PON. Where the ONT mounts, inside or out, and how the drop enters the building is the install detail that determines whether the customer experience is clean or a callback.

The drop is also where OSP volume lives. Feeder and distribution get built once, but drops get installed one home at a time for years as customers sign up, so the drop process, the slack at the tap, the connector or splice at the home, and the as-built of which port serves which address, is the part of the plant that a crew touches most often.

How do you test an OSP fiber link?

You prove an OSP link two ways, and a finished link needs both. An end-to-end insertion loss test with a light source and a power meter measures the total loss of the link against the optical loss budget: source to detector, including every splice and connector. An OTDR trace shoots a pulse down the fiber and reads the reflections back, building a picture of the link that locates each splice, each connector, each bend, and each break by distance and by loss. The power meter tells you whether the link passes. The OTDR tells you why and where.

Single-mode OSP is tested bidirectionally on the OTDR, meaning from both ends, and the two directions are averaged. The reason is real: an OTDR can read a splice between two slightly different fibers as a gain in one direction and a loss in the other, and only the average of both directions gives the true splice loss. A one-direction trace will pass a bad splice or fail a good one depending on which way you shot it.

Acceptance is against the loss budget and the splice-loss limits in the project spec and the TIA testing standards. The link must come in under the calculated budget for its length, splits, and connector count, with each splice under its limit. Document both the power-meter loss and the OTDR traces, because the trace is the baseline a future tech compares against when the link degrades. Verify the test method and acceptance values against the TIA standards and the engineer's loss budget.

The OTDR and the optical loss budget

The optical loss budget is the math that says a link will work before you build it and the pass/fail line after. You add up the expected loss along the path: the fiber loss per kilometer times the length, plus a loss allowance for each fusion splice, each connector, and each splitter with its split ratio. The transmitter has so much power and the receiver needs a minimum, and the difference is the budget the link has to stay under.

The split is the big number in an FTTH budget. A passive splitter spends a large, fixed chunk of the budget, more for a higher split ratio, which is why the split plan and the reach are linked. Run the budget and a 1 to 64 split simply will not reach as far as a 1 to 32 on the same fiber and optics.

The OTDR is how you see the budget realized in the actual glass. It plots loss against distance so you can see a high splice, a stressed bend losing light, or a break, and read the distance to it within the instrument's resolution. The trick the OTDR cannot beat is its dead zone near the connectors, which is why a launch fiber goes on the front and a receive fiber on the back, so the first and last real events fall in the readable part of the trace, not in the blind spot at the bulkhead.

The as-built and GIS mapping

The as-built is the record of where the fiber actually is, and on an OSP network it is as important as the cable. A buried network is invisible. If the records do not show where the route runs, where the splices and slack sit, which strand a fiber is on, and which port serves which home, then the next crew is locating and digging blind on a network they own. The as-built is what makes the plant maintainable and what makes the next phase buildable.

Modern OSP records live in GIS. The route, the poles, the handholes and vaults, the splice closures, the splitters, the slack loops, and the fiber assignments get captured as mapped, attributed features tied to real coordinates, so a tech can pull up the map and know what is under the street before the shovel comes out. The as-built captured during construction, redlined against the design where the field had to deviate, is what feeds that map.

The failure mode is the build that gets placed and never recorded, or recorded weeks later from memory. The route shifted around a rock, the splice moved to the next pole, the count changed, and none of it made the map. A field tool that captures the as-built where the work happens, the locate ticket, the closure location, the splice record, the photos, the OTDR trace, geotagged at the point of work, keeps the map honest. FieldOS is built to capture that field record as the crew works, so the as-built reflects what got built, not what got drawn.

Permits, right-of-way, and easements

OSP construction happens on land you do not own, so the legal right to be there comes before the cable. Public roads carry a right-of-way (ROW) the road authority controls, and crossing or running along it needs a permit from the DOT, the county, or the city, with conditions on depth, restoration, and traffic control. Railroad crossings are their own world, with the railroad's own permit, insurance, and flagging requirements, and they take longer than anyone plans for. Private land needs an easement, a recorded legal right to place and maintain the cable.

On poles, the right to attach comes through the pole owner's joint-use agreement and the make-ready process, which is its own permit in everything but name. None of this is fast, and the permitting timeline often sets the construction schedule more than the build does.

The field consequence is that the crew cannot outrun the paper. Place cable in a ROW without the permit or on private land without the easement and the work can be ordered out, restored, and re-permitted at the contractor's cost. Confirm the permit, the ROW, and the easements are in hand and that the crew is working inside their conditions before mobilizing, because the road authority, the railroad, and the pole owner each set their own rules.

Restoration: putting the surface back

Restoration is the part of the underground build the public actually sees, and a build is not done until the surface is back. The trench gets compacted and backfilled in lifts so it does not settle into a trough a year later. Pavement gets patched to the road authority's spec, which usually means a saw-cut edge, the right base, and a finished surface that matches, not a cold-patch lump. Bore pits, handhole excavations, and the lawn across a buried drop all get put back.

The standard is set by the permit, and the road authority inspects against it. A settled trench, a failing patch, or a lawn left rutted is a callback and a black mark on the next permit application from the same contractor. Restoration that fails its inspection gets redone at the contractor's cost.

The crew lesson is that restoration is planned and priced up front, not scraped together at the end. The compaction, the patch detail, and the landscape repair are line items, and the build that skimps on them pays twice.

Safety: traffic, trench, pole, and laser

OSP safety covers more hazards than inside work, because the job is outdoors in public space. Traffic is the first one. Crews work in and beside live roadways, so a work zone with the proper signs, cones, and flagging is set up before anyone steps into the road, per the traffic-control plan for the permit. Get hit by a car and the fiber does not matter.

The ground and the poles carry their own hazards. An open trench deep enough to enter is a trench-safety situation with cave-in risk, requiring the protection the excavation rules demand; the trench-safety detail is shared with any excavation trade. Poles mean working at height in a bucket or on climbers, near energized power above the communication space, which is exactly why the NESC clearance exists and why you stay in the communication space.

Fiber has a hazard people forget: the light. The laser in a live fiber is invisible and can damage your eye, so you never look into the end of a fiber or a connector that might be live, and you treat every fiber as live until proven dark. Glass scraps from cleaving are sharp and get into skin and clothing, so they go in a dedicated container, never on the ground. And the dig hazard loops back to the locate: the safest excavation is the one that started with a good 811 ticket and verified marks.

What to document

An OSP build that is not documented is a network nobody can maintain. The record is what the next crew, the locator, and the engineer rely on years later, and on a buried plant it is the only thing that says where the fiber is.

Capture the locate ticket and its positive responses, the route as built with deviations from the design redlined, the pole attachments and make-ready, the burial depth and the warning tape and tracer, the closures and access points with their locations and slack lengths, the splice records, the OTDR traces and the power-meter loss for each link, and the permits and easements. Photos geotagged at the point of work tie it all to the ground. Feed it to the GIS so the map matches the plant.

Common mistakes

• Digging without an 811 locate, or after the ticket expired, and hitting a gas, power, or fiber line.
• Violating the NESC clearance on a pole by crowding into the power space or shorting the safety zone.
• Placing cable that is not water-blocked, or leaving a closure unsealed, so water gets in and freezes.
• Bad fusion splices from a worn cleaver or dirty prep, passing a high-loss splice the OTDR would have caught.
• Leaving no slack at closures and access points, so the next repair needs a new cable section and more splices.
• Building with no as-built or GIS record, leaving a buried network nobody can locate or maintain.
• Drilling a cross-bore in HDD through a sewer lateral, the strike that surfaces years later as an explosion.
• Testing the OTDR one direction only, which hides bad splices and fails good ones depending on direction.

Field checklist

Standards and references

Three bodies of rules govern OSP, and each controls a different part of the job. Damage prevention runs through state one-call law and the 811 system: you locate before you dig, you honor the notice period and positive response, and the marks and tolerance zone control how you excavate near a line. This is statute, and it is the one rule with no exceptions.

Aerial construction runs to the National Electrical Safety Code (NESC). The NESC sets the supply and communication spaces on the pole, the safety separation between power and communications, the ground clearances by what the span crosses, and the loading the strand has to carry. The pole owner's construction standards and joint-use agreement layer on top of the NESC and are usually stricter, so confirm both, and confirm the adopted edition.

Fiber, splicing, and testing run to the TIA fiber standards and the manufacturer's specifications. The TIA documents and the test procedures cover insertion-loss and OTDR testing, acceptance values, and the loss budget approach; the cable and closure manufacturers set splice-loss limits, bend radius, and sealing. The engineer's design and loss budget control the numbers for the specific build. The plant lives or dies on three habits: locate before you dig because 811 is the law, choose aerial or underground and place water-blocked cable to the NESC, and fusion-splice, OTDR-test, and map the as-built. Verify every clearance, depth, and loss value against 811, the NESC, the TIA standards, the engineer, and the AHJ and pole owner, because the controlling numbers depend on the jurisdiction and the design.

Units, terms, and acronyms

OSP carries its own vocabulary, and the same idea shows up under several names across a design set, a manufacturer sheet, and a permit. These are the terms a crew and an inspector use on the job.

Loss is in decibels (dB), distance in feet or kilometers, burial depth in inches, and the optical loss budget is the total dB a link is allowed to lose end to end. Split ratio is written as 1 to 32 or 1 to 64, the number of homes one feeder fiber serves through a splitter.

Outside plant (OSP)

The outdoor part of the network, buried or aerial, from the central office out to homes and businesses

FTTH / FTTx

Fiber to the home, and the family of variants (curb, node, building) where fiber stops short of the home

811 locate / damage prevention

The one-call system and the law requiring utilities to be located and marked before any dig

Strand / lashing

The steel messenger that carries aerial load, and the wire that lashes the fiber cable to it

HDD / microtrenching

Horizontal directional drilling steers a bore underground; microtrenching cuts a narrow shallow slot in pavement

Loose-tube / ribbon cable

Fibers floating in buffer tubes, or bonded into ribbons for high count and mass fusion splicing

Fusion splice / closure

Two fibers arc-welded into one, protected and housed in a sealed re-enterable enclosure

PON / OLT / splitter / ONT

Passive optical network: the head-end OLT, the passive splitter, and the ONT at the home

OTDR / loss budget

The instrument that maps splices and faults by distance, and the total allowed link loss in dB

Related tools

Calculators and readiness checks for this work

Compare your options

FAQ

People also ask

• What is a low-voltage system?
• What is the difference between single-mode and multimode fiber?
• What is a fiber loss budget?
• What is an assured equipment grounding conductor program (AEGCP)?
• What is involved in cell tower construction?