Copper networks have a shelf life. The engineering question isn't whether to migrate — it's how to do it without tearing up your service territory twice.
The operators we see struggle the most with migration aren't the ones who lack capital — they're the ones who started construction before the engineering was finished. A copper plant that's been in the ground for 23 years doesn't give up its topology easily. Serving area interfaces, buried duct banks, DLC huts, and aerial strand that's been lashed to the same poles since before the internet existed — all of it has to be mapped, reconciled, and accounted for before a fiber overlay route makes any sense.
Here's a scenario that's not uncommon: a rural ISP in the Southeast decides to build fiber over their copper territory. They hire a contractor to design the overlay. The contractor draws routes in QGIS without doing a field survey of the existing copper plant first. Six months into construction, crews hit a duct bank that wasn't on any GIS layer — 847 copper pairs running through a 4-inch conduit right where the fiber route was supposed to go. They can't trench parallel to it without violating separation requirements. The route has to move. That move adds $1,340/mile in incremental construction cost and pushes the schedule four months. That's not a contractor failure — it's an engineering failure that happened before the contractor touched a shovel.
Migration also creates service continuity risk that greenfield fiber builds don't have. You're not building to empty addresses — you're replacing the active service that subscribers are paying for right now. A cutover sequence that doesn't account for which serving area interfaces feed which address ranges will interrupt service. We've seen ISPs accidentally cut over two SAIs simultaneously because the splitter architecture didn't match the copper serving area boundaries, dropping 340 subscribers who were still on copper. That's a planning failure. It's the kind of thing that proper copper to fiber migration engineering prevents.
The OSP design discipline required for copper retirement is different from greenfield fiber. You need someone who understands both the legacy copper plant — pair counts, DSL serving areas, DSLAM removal logistics — and the fiber network that's replacing it. Designing an FTTH overlay without understanding what's underneath it is how you end up designing the same territory twice.
Our process runs from field-verified copper plant audit through phased cutover engineering and legacy decommission documentation. We don't design the fiber network until we know the copper network — that's not optional, it's the sequence that prevents the mistakes described above. Typically 8–14 weeks for the engineering phase on a rural ISP project covering 1,200–2,500 addresses.
We start with a full audit of the copper plant — pair counts by serving area, SAI locations, DLC hut inventory, buried vs. aerial route breakdown, and conduit reuse potential. Our field crews use Katapult to GPS-tag and photo-document the plant; that data feeds directly into QGIS for route conflict analysis. We're not working from 1990s OSS records that may or may not reflect what's actually in the ground. We verify.
With the copper plant mapped, we design the fiber overlay — distribution routes, feeder routes, and drop design — in AutoCAD and QGIS, routing around existing buried copper and duct banks to avoid conflicts. Route design includes aerial attachment planning with O-Calc Pro clearance checks, permit drawing preparation, and home passings reconciliation against the operator's address list. We flag every location where the fiber route intersects existing copper infrastructure.
We sequence the migration by serving area — typically SAI by SAI — to minimize the number of addresses at risk during any single cutover window. The phase plan defines which SAI feeds which fiber distribution area, the subscriber count per phase, and the engineering dependencies between phases. This is where the FTTH LLD splice point placement decisions get made to match the phasing boundaries.
Cutover engineering specifies the exact sequence for each address: fiber ONT installation, service activation on the fiber network, subscriber verification, and copper pair disconnection. We produce the cutover work order package for each phase — including fallback procedures if fiber activation fails and the subscriber needs to remain on copper temporarily. The goal is zero unplanned service interruptions. It's achievable. It requires a written plan.
Once cutover is complete, we produce the decommission package — as-built records of copper removed or abandoned in place, splice point documentation, conduit reuse mapping, and MDF/IDF transition drawings. DSLAM removal logistics are coordinated with the equipment team. Buried copper conduit that's being repurposed for fiber gets documented with condition notes and cross-section dimensions. This documentation is what your operations team needs for long-term network management after the copper is gone.
The full engineering scope we deliver covers every phase of the migration — not just the fiber network design, but the audit work that makes the fiber design accurate and the cutover planning that makes the migration safe.
Field-verified inventory of the existing plant — pair counts, SAI locations, DLC sites, buried cable routes, aerial strand condition, and duct bank mapping. We don't design around what we think is there. We design around what's confirmed.
Full fiber network design — feeder, distribution, drop — in AutoCAD and QGIS, with route conflict analysis against the existing copper plant. Includes permit drawings and make-ready engineering for aerial routes.
We reconcile the operator's address list against the field-verified copper serving area topology to confirm every address is accounted for in the fiber design. Addresses that fall outside standard drop length get flagged for engineering review.
GPON or XGS-PON splitter ratio design — typically 1:32 or 1:64 depending on subscriber density — with split point locations designed to align with the phased migration boundaries and existing copper SAI coverage areas.
OLT placement, ONT specification, and MDF/IDF transition design. We specify where active electronics go, what happens to the DSLAM equipment during cutover, and how the headend transitions from copper to fiber-based subscriber management.
Post-construction as-built documentation — fiber routes, splice points, conduit reuse, and decommissioned copper inventory. The records your NOC needs to manage the network after migration is complete.
You can't design a fiber overlay correctly without understanding the copper network it's replacing. Most of the expensive surprises in copper retirement projects — route conflicts, address coverage gaps, cutover sequencing errors — trace back to inadequate plant assessment before design began. Here's what we actually look at.
This audit phase — done with Katapult in the field and QGIS in the office — is what the fiber overlay design depends on. Skipping it or abbreviating it to save time upfront is one of the most reliable ways to add cost and schedule to the back end of the project. We've detailed the common planning failures in our guide to FTTH HLD design mistakes — many of them apply directly to copper retirement projects.
There are two broad strategies for copper retirement, and the right one depends on your existing plant topology, subscriber density, capital budget, and BEAD eligibility requirements. We've designed both — here's how we think about the tradeoff.
Fiber overlay (parallel build): You build the fiber network while copper remains active, migrate subscribers one by one, and retire copper after each serving area reaches full fiber penetration. This approach keeps service continuity risk low — a subscriber is never without service during the migration — but it requires operating two parallel networks during the transition period. That has real operational cost. Overlay also lets you phase capital investment over 2–4 years instead of front-loading it all. It's the right approach for most rural ISPs with existing copper infrastructure and active subscribers.
Brownfield replacement (hard cutover): You build fiber, cut all subscribers in a serving area simultaneously on a scheduled maintenance window, and retire the copper in the same cycle. The operational cost of dual-network operation goes away. But the cutover risk is higher — if something goes wrong during activation, you've got a large block of subscribers offline. Hard cutover also requires more preparation: every ONT has to be pre-installed and tested before you pull the copper circuit. We've seen this work well on serving areas under 400 addresses with high-confidence fiber activation records. We don't recommend it for first-time migrations where the team doesn't have fiber activation experience yet.
The hybrid approach — overlay construction, SAI-by-SAI phased cutover — is where most of our projects land. It captures most of the risk reduction benefit of overlay while compressing the dual-network operation period by migrating serving areas in batches rather than address by address. See also our guide on fiber network design outsourcing for context on how the engineering phases integrate with construction management.
| Factor | Greenfield FTTH | Brownfield Overlay | Hybrid (Phased SAI) |
|---|---|---|---|
| Cost per mile (engineering + construction) | $28K–$45K (no copper conflicts) | $34K–$58K (conflict avoidance, higher complexity) | $31K–$52K (phased, amortized) |
| Service disruption risk | None (no active subscribers) | High during hard cutover window | Low — one SAI at a time |
| BEAD eligibility | Eligible if unserved/underserved | Eligible if copper fails speed threshold | Eligible — most common BEAD scenario |
| Timeline to first revenue | 18–36 months (full buildout) | 12–18 months (faster single-phase) | 6–12 months (first phases live early) |
| Dual-network operation period | Not applicable | Minimal (weeks) | 12–36 months during phased rollout |
| Engineering complexity | Moderate | High (conflict avoidance, DSLAM logistics) | High (phasing sequencing, cutover engineering) |
Cost benchmarks for copper retirement projects vary significantly by geography, plant type, and subscriber density. These are real-world ranges from projects we've engineered — not estimates from a spreadsheet model.
| Cost Category | Typical Range | Notes |
|---|---|---|
| OSP engineering per mile | $1,200–$2,800/mile | Higher end for brownfield with significant conflict avoidance engineering |
| Aerial construction per mile | $18,000–$32,000/mile | Includes make-ready; varies by pole density and make-ready rate |
| Underground construction per mile | $38,000–$85,000/mile | Wide range driven by soil conditions and bore vs. open trench |
| Drop installation per address | $350–$800/address | Aerial drops on low end; longer buried drops on high end |
| Cutover engineering per SAI | $1,800–$4,500/SAI | Includes cutover work order package and fallback procedures |
| DSLAM removal and decommission | $2,200–$6,000/site | Varies by equipment age, vendor, and hut size |
| MDF/IDF transition per CO | $4,500–$12,000/CO | Higher for large central offices with complex frame cross-connections |
Engineering cost is typically 8–14% of total project cost on copper retirement projects — higher than greenfield because the audit and conflict analysis work doesn't exist on a clean-slate build. The 3.4 dB insertion loss budget you're designing to on an XGS-PON overlay also requires more careful splitter placement analysis than a simple greenfield GPON deployment. The last mile fiber design discipline matters in brownfield more than anywhere else — there's less room for error when you're threading fiber through a territory that's already full of plant.
FREE FIRST PROJECT
Active in 22 states. First 20,000 LF of copper-to-fiber migration design free — no commitment required. Covers a full SAI serving area evaluation for most rural ISPs.
Start Free Design →The central office and remote terminal side of a copper retirement gets less attention than it deserves in most migration plans. You're not just running fiber to the subscriber — you're decommissioning the circuit that runs from the subscriber pair back to the DSLAM port at the CO or remote DLC hut. That requires a plan.
MDF transition starts with a cross-connect audit. Every active pair that's being migrated to fiber has to be located on the MDF, verified against the subscriber record, and scheduled for disconnection after the ONT activation is confirmed. On a 23-year-old MDF with multiple generations of OSS records, the cross-connect documentation frequently doesn't match the physical frame. We've found abandoned pairs still carrying DSL signal from DSLAMs that were supposed to have been decommissioned two years earlier — the port was never administratively disabled, so the DSLAM kept trying to sync. That's the kind of thing a physical audit catches.
IDF transitions at remote DLC sites follow the same discipline — inventory the active line cards, map each port to a subscriber record and a copper pair, and sequence the decommission alongside the fiber cutover for that serving area. We produce the IDF transition drawings as part of the cutover engineering package, with before-and-after frame diagrams for each CO and remote site involved in the migration.
For GPON network design specifics — OLT port planning, ONT provisioning sequencing, and how the active fiber electronics integrate with the headend — we cover that in depth in the GPON design guide.
Copper to fiber migration engineering covers the full scope from existing network audit through legacy plant decommissioning. That means field survey of the existing copper plant — pair counts, serving area interfaces, buried cable routes, DLC locations — followed by fiber overlay route design in QGIS and AutoCAD, splitter architecture, home passings reconciliation, phased cutover planning, and permit drawings for ROW and make-ready. We also produce equipment location plans for OLT/ONT placement and coordinate MDF/IDF transition scope for each serving area. The field survey and the fiber design are engineered together — not handed off sequentially to different teams.
Timeline depends heavily on serving area size, aerial vs. buried construction, and whether you're doing a phased rollout or a hard cutover. A single serving area of 1,200 addresses typically runs 4–7 months from field survey to service cutover — assuming permitting doesn't hit major delays. Larger projects with multiple SAIs and DLC sites, or those requiring duct bank coordination with existing buried copper, run 12–24 months. The engineering phase alone — network audit, fiber overlay design, permit drawings — is 8–14 weeks for most rural ISP projects. Don't compress it.
Yes, with proper phased migration planning. The standard approach is fiber overlay — build the FTTH network parallel to the copper plant, migrate subscribers address by address, and cut copper after fiber service is confirmed active and stable. That way you're never dropping an active subscriber's service to execute a cutover. The engineering challenge is designing the overlay to avoid route conflicts with existing buried copper and duct banks, and sequencing the SAI-level cutovers so each serving area completes cleanly before the next begins. We write the cutover work order package for each phase, including fallback procedures if something doesn't activate correctly.
We produce legacy plant decommission documentation as the final deliverable of the migration — as-built records of copper removed or abandoned in place, splice point documentation, conduit reuse mapping, and MDF/IDF transition drawings. Many ISPs leave buried copper in place rather than pulling it — the cost of removal typically isn't justified unless the conduit has reuse value or the ROW agreement requires restoration. Buried copper conduit being repurposed for fiber gets documented with condition notes and cross-section dimensions. DSLAM removal logistics are coordinated as part of the IDF transition scope.
BEAD funding covers deployment to unserved and underserved locations — and copper retirement projects in rural areas frequently qualify when the existing copper plant doesn't meet the 100 Mbps/20 Mbps threshold that defines "underserved" under NTIA's rules. The engineering documentation required for BEAD applications — area maps, address-level passings counts, construction cost estimates, and network design narratives — is the same work we produce as part of the migration engineering package. We've supported BEAD subgrantee applications across multiple states. If your territory qualifies, we know what the documentation requirements look like.
Both. Most copper retirement projects we've engineered go GPON — specifically XGS-PON for new builds — because the passive splitter architecture eliminates active electronics in the field and reduces long-term maintenance costs compared to a copper DSL plant. Active Ethernet is sometimes specified by clients who need guaranteed symmetrical bandwidth per subscriber, particularly in commercial-heavy service territories. We design the splitter architecture, fiber counts, and OLT placement for either platform. Our GPON network design guide covers the design methodology in depth.
MDF transition starts with a cross-connect audit — locating every active pair on the MDF, verifying it against the subscriber record, and scheduling disconnection after ONT activation is confirmed. IDF transitions at remote DLC sites follow the same discipline: inventory active line cards, map each port to a subscriber record and copper pair, and sequence the decommission alongside the fiber cutover for that serving area. We produce before-and-after frame diagrams for each CO and remote site involved in the migration. The cutover work order package includes the MDF/IDF transition steps for each phase. It's not glamorous work, but it's where mistakes get made if the plan's not written down.
We're active in 22 states and available to deploy across all 50 U.S. states for copper to fiber migration engineering. Our highest-volume markets for copper retirement projects include Florida, Texas, North Carolina, Georgia, Ohio, Virginia, and Pennsylvania — but we've staffed projects from Alaska to Maine. Whether you're a small rural telephone co-op retiring 800 copper pairs or a Tier-2 CLEC migrating 85,000 subscribers to fiber, we can scope and staff the engineering. Contact us at info@draftech.com with your project details.
ARE YOU A COPPER-TO-FIBER MIGRATION FIRM?
This page describes the service we deliver to clients. If you provide copper retirement engineering or FTTH overlay design and you're looking for a consistent subcontract pipeline, we have ongoing capacity needs in this discipline.
Tell us your service territory size, approximate address count, and whether your existing plant is primarily aerial or buried. We'll scope the engineering and give you a timeline estimate. We've done this across 22 active states — if there's a plant type or permitting jurisdiction we haven't seen before, we'll tell you that too.
Contact Our Engineering TeamOr email directly: info@draftech.com — we reply within one business day. First 20,000 LF at no charge: draftech.com/free-design
SERVICE AREAS
Active in 22 states and deployable across all 50 U.S. states — including our highest-volume BEAD markets:
View all service areas →Draftech International provides copper to fiber migration engineering services across all 50 U.S. states — from small telephone co-ops retiring aging copper plant to Tier-1 carriers and BEAD-funded subgrantees. Contact our engineering team to discuss your project.