Fiber Strand Count Design Planning: Take Rate, Layers, and Standard Count Steps

A BOM that orders one cable size for every layer is how a 24-count lateral gets printed on a feeder that still has to serve four cabinets. The crew then opens a sheath that was never meant to be accessed at that span. Fibers the owner thought were parked are already assigned.

This guide is the count worksheet. Take rate at the growth horizon comes first. Each layer then gets its own row, and spare is named before we round to a standard 12-fiber step. Split-ratio math lives in our FDH sizing post. Cable construction families live in their own purchasing matrix. Fiber strand here means a glass fiber inside the sheath. It does not mean messenger wire.

What Fiber Strand Count Design Planning Sizes

Fiber strand count design planning is the worksheet that turns passings, take rate, split, and a named spare into a cable count on each layer, then rounds that demand to a standard 12-fiber step. We size each layer on its own row. A 10-year horizon is the default we record unless the owner writes a different one.

Fiber Strand Count Design Planning Worksheet

The worksheet is low-level design of capacity, not a purchasing catalog. ITU-T G.984.1 and G.9807.1 describe PON systems; neither recommendation picks a street-cable fiber count. The owner still has to map optical interfaces onto physical plant.

Count is also not a role. A 288-fiber sheath can carry backbone assignments on some fibers and feeder assignments on others. Our fiber backbone vs distribution design guide keeps those roles on the drawings. This sheet only answers how many fibers each role needs on that span.

Draftech's in-house fiber network design writes the worksheet into the LLD. Engineering stays in-house. When construction is in scope, Draftech-managed subcontract crews deliver full turnkey under our QA/QC and safety program. We do not claim a self-perform construction crew today.

The difference between HLD and LLD matters here. HLD can carry a planning count. LLD is where the count is frozen against actual routes and splice locations. It is also frozen against the cable the mill will ship. Ordering from an HLD round number is how a 144 becomes a 96 at the yard with no one updating GIS.

Start With Take Rate and the Growth Horizon

Passings are not subscribers. Take rate is the share of those passings expected to take service at the growth horizon the owner approved. Year-one marketing is the wrong input because it understates the fibers the mill has to ship on the feeder row against the 10-year take the owner already approved. We apply take rate before we pick a cable size, because a 400-passing area at 60 percent take is 240 live paths, not 400.

Our published FDH method uses a 10-year horizon as the default. The take rate on the worksheet comes from the owner's approved market forecast for that serving area, not from a universal market range. The same approved value belongs on the cable and port worksheets, or the feeder count and cabinet ports will disagree.

Split is an input to this sheet, not the work of this sheet. The split ratio converts expected subscribers into feeder fibers and FDH ports. Optical-budget limits on 1:16, 1:32, and 1:64 live in our fiber distribution hub sizing guide, including the 85 percent port-utilization ceiling. We take the approved split as a given here and refuse to invent a second one on the cable row.

Dedicated circuits do not share that math. A wireless backhaul pair, an enterprise Ethernet handoff, or a dark-fiber commitment uses assigned fibers, not a PON split. We put those assignments on their own line so a residential take-rate model cannot swallow them. If the owner has not approved the horizon for those circuits, we hold the count.

The table is a preview map. It does not require five separate cable families on every span. One sheath can carry more than one row. We still keep the rows, because a spare fiber with no pool name is not a spare. It is a rumor.

Size Each Layer From Backbone to Drop

Each layer answers a different question. Mixing them into one total is how a neighborhood lateral inherits a backbone spare rule, or how a drop count gets used to size a feeder. We walk the path from the core toward the premises so the numbers stay attached to the plant they actually occupy.

Backbone count is aggregation, not a neighborhood total

Backbone demand starts with the services that leave a major node. That includes aggregated access toward several serving areas, plus any circuit that cannot ride a PON. Diversity, where the operating plan requires it, is a second assignment, not a percentage sprinkled on the first. We do not double a count and call the extra copy spare. If a protected path needs its own fibers, those fibers get their own line and their own route check.

A river crossing or a long conduit run that will be expensive to reopen can justify backbone reserve for a territory that is not in the first phase. That reserve still needs a named horizon. Five years of unknown growth is not a reason. An owner-approved second cabinet area on the far side of that crossing is.

Feeder count follows ports, then the cable

Feeder demand is the port plan after take rate and the approved split, plus any dedicated feeder assignments. The FDH article owns that port math. What this row still has to do is convert those ports into fibers on the cable that actually reaches the cabinet, then add the cable spare the operator can use without opening the splitter tray.

Feeder size is an outcome, not a starting point. An isolated rural cluster whose passing count is capped by geography may land on a smaller stock count than an urban serving area or a route with dedicated protection assignments. Copying a larger feeder count onto that cluster is a warehouse problem, not a safety factor, because the extra fibers sit unpaid in a sheath that geography will never fill at the approved take rate.

Distribution and drop are local, and they stay local

Distribution demand is the terminal layout. A 12-count lateral that serves eight homes with a 1:8 distributed split is a different row from a 48-count branch that still has to pick up two future streets. We size the branch from the terminals it actually reaches, then add only the redevelopment reserve the owner named for that serving area. A backbone spare rule does not belong on that branch.

Drop count is usually 1 live fiber, or 1 live plus 1 spare, on a single-family drop. MDU laterals often step to a 12-count because the color code and the splice tray are already built that way. We do not invent a 6-count drop cable to match a unit count of 5. The mill will not thank you, and neither will the technician who has to remember a one-off palette.

Name the Spare Before You Add It

Spare is not a vibe. A single 20 percent printed on the title block and applied to every layer is how maintenance fibers get consumed by a late subdivision, and how a protection pair disappears into a round number. We name the pool on the same row as the count.

One thing I still catch in our reviews: that title-block percent. I have signed packages that carried it. That limitation is on us: the operator then asked which of the unused fibers they were allowed to cut over, and the drawing had no answer. We now refuse a spare line that cannot name its job.

• Maintenance pool: fibers the operator can use for restoration or a failed drop without touching committed growth.
• Committed growth: capacity the owner already approved for a named horizon, recorded as assigned rather than leftover.
• Discretionary reserve: fibers that can be released if a development is cancelled, with an owner who can say so.
• Exhaust trigger: the threshold that starts augmentation design before the available count hits zero.

Those four lines are the spare policy. They are not interchangeable. A fiber reserved for a known second phase is not available to a technician on a Saturday cut. GIS has to carry the same pool name on the as-built, or the next designer who opens the sheath two years later will treat every dark fiber as free capacity and assign it to a Saturday cut.

The 85 percent FDH utilization ceiling is a port spare, not a cable spare. Mixing the two is how a cabinet still has empty ports while the feeder sheath is full. We keep them on separate rows and reconcile them before the cable is ordered.

Round to Standard Count Steps

Demand is a calculated number. The cable you order is a stock step confirmed against the selected manufacturer's current catalog. Many OSP cable families use 12-fiber color-code groupings, but available constructions and counts still have to be checked before release. The worksheet records the calculated demand and the next available stock count rather than inventing a custom count.

We round after spare is applied, not before. Rounding a 61-fiber demand to 72 and then adding 20 percent produces 86.4 fibers before whole-fiber and stock-count rounding, forcing another step. That is two rounds of inflation. Apply the named pools first. Then step up to the next stock count. Write both numbers on the row so a reviewer can see the calculated demand and the ordered count as two cells, not one.

Cable construction is a later decision. Loose-tube, ribbon, and microcable families, plus glass and jacket types, sit in the OSP cable-selection matrix we already publish. This worksheet freezes how many fibers. It does not pick the sheath. Substituting a ribbon 144 for a loose-tube 144 does not reopen the count unless the owner also changes the splice model.

A special-count cable that no mill stocks is a delay wearing an engineering costume. The yard will substitute. GIS will not. We would rather order 144 and park 12 fibers in a named pool than pretend a 132 will arrive on the date the crew is scheduled.

Count Recommendations by Owner Type

Rural cooperative or single-state ISP under 5,000 passings: freeze take rate and the 10-year horizon before anyone sizes a feeder. Use the approved FDH port plan, then size the cable from those ports and named dedicated assignments. Select the next available stock count from the calculated row rather than copying a larger serving area's feeder onto a small pocket.

Overbuilder on a 2026 BEAD schedule: lock the count model at LLD, not at the HLD planning number. Grant closeout will ask what was installed. A substitution from 144 to 96 that never hit GIS is a records failure. Hold the cable order until the named spare pools and the standard step are on the same row. Segment-based release lets one serving area order while another is still in review.

Campus or municipal telecom owner with dedicated circuits: keep PON take-rate math off the enterprise and wireless rows. Assign those fibers by circuit. Round each row to a 12-fiber step on its own. A 12-count that carries eight live campus links and four named spares is clearer than a 24-count that nobody can explain two years later.

The useful outcome is not a bigger cable. It is a worksheet that shows calculated demand, named spare, and the stock step actually ordered, with each layer on its own row. An operator who did not design the plant should be able to find a parked fiber without calling us.

That is the problem this service removes. A single count printed across every layer is the first failure. A title-block spare percent with no pool name is next, and it is the same failure that leaves an operator unable to say which unused fiber they may cut over on a Saturday restoration. We also see custom mill counts that never ship, plus feeders sized from drop math. Draftech keeps the count model in-house so those four failures stay on the sheet, where they can still be fixed.

Active in 24 states. Available across all 50 U.S. states. If a live route needs the count worksheet before cable is ordered, start with a free 20,000-ft design review. For a package already in draft, reach out at info@draftech.com.