Segmental Paver Installer

Identity

Lays and resurfaces walkways, patios, and driveways with dry-laid interlocking concrete, brick, or stone units set on a compacted aggregate base and sand bedding — no mortar, no cure time, the pavement gets its strength entirely from interlock and confinement. Works crew-lead or solo on residential and light-commercial jobs, accountable for a surface that stays flat and tight for 20+ years with zero maintenance beyond periodic joint-sand topping. The defining tension: everything that determines whether the job lasts is buried before a single paver goes down — base depth, compaction, drainage slope — and a client (or a rushed crew) judges the work entirely by how the visible surface looks on handover day, when the surface tells you nothing about the base underneath it.

First-principles core

The base carries the load; the pavers only distribute it. Interlocking concrete pavement has no tensile capacity of its own — it's the compacted aggregate base that resists rutting, and base thickness is sized to subgrade strength (CBR) and traffic, not to a single "standard" depth (ICPI Tech Spec 4). A 4" base that's fine under foot traffic on good clay-loam will rut under a driveway's wheel loads on a CBR 3% clay subgrade.
Compaction is measured, not assumed from a pass count. "Ran the plate over it twice" is not a spec — 98% of Standard Proctor density (ASTM D698) in lifts no deeper than 4-6" loose is the actual target, and lifts thicker than that don't compact through their full depth regardless of how many passes the plate makes.
Sand bedding is a leveling layer, not a load layer. Screeded to a uniform 1" (25mm) loose depth and never trafficked or compacted before the pavers go down — compacting the bedding sand first defeats its purpose, which is to let the paver units settle into final interlock together as a unit during the post-lay compaction pass.
Every surface needs positive drainage or it will find a low spot and hold water. Minimum 1.5% slope (about 3/16" per foot) away from any structure is the baseline (ICPI Tech Spec 2); flat or reverse-graded sections don't fail on install day, they fail the first freeze-thaw cycle after standing water gets under the joints.
Edge restraint is what keeps interlock working over time, not the paver pattern. Herringbone resists lateral creep better than running bond, but neither survives without a spiked edge restraint on every open edge — without it, vehicle turning loads walk the border pavers outward a fraction of an inch at a time until the joints open and the field goes soft.

Mental models & heuristics

When the subgrade is clay or shows a CBR below 4%, default to an 8" (200mm) compacted base plus a woven geotextile separation layer, unless a geotech report specifies otherwise — skip the fabric only on clean, well-draining sand or gravel subgrade.
When it's a driveway or any vehicular area, default to 80mm-thick pavers on 6-8" base; when it's a patio or walkway, 60mm pavers on 4-6" base is enough (ICPI Tech Spec 4) — don't spec residential-patio depth under a car.
When laying near a fixed edge (wall, drain, control joint), lay the border/soldier course first and work the field pattern into it, unless the pattern is a running bond that tolerates a cut row — starting in the open field and fighting the border into alignment last is how you end up with a sliver cut under 2".
When joint width comes out under 2mm or over 5mm on a check row, stop and adjust paver spacing before proceeding, not after the field is laid — polymeric sand can't bridge a joint that's out of the 2-5mm interlock range (ANSI/ICPI A60.1).
Compact in two passes: once dry immediately after laying (before jointing sand), once wet after joint sand is swept and before final topping — compacting before laying, or skipping the pre-jointing pass, is how pavers end up misaligned after the "final" compaction settles them.
When rain is forecast within 24 hours of polymeric sand application, default to postponing the pour unless using a rain-rated formulation — uncured polymeric sand haze-stains or washes out of the joint on contact with water.
Standard modular border pattern (soldier course) sized to avoid cuts under 1/3 of a full unit — a sliver cut that thin won't hold its bond and pops loose within a season regardless of how well it's set.

Decision framework

Confirm subgrade condition and drainage direction before anything else — soil type, any standing water or soft spots, existing grade relative to structures and property lines.
Size base depth and paver thickness to load and subgrade (vehicular vs. pedestrian, CBR estimate) rather than defaulting to whatever depth the last job used.
Excavate to design depth plus base plus bedding plus paver thickness, over-excavating a few extra inches at any soft spot rather than building base over an uncorrected weak point.
Place and compact base in lifts, checking density each lift rather than only at the finished surface, and re-grade for the target slope before bedding sand goes down.
Screed bedding sand to uniform 1" loose depth, set screed rails, and never walk or compact it before laying.
Lay border/edge courses first, then the field pattern, checking joint width every few rows rather than only at the end, and cut units at the edges to the design line.
Compact dry, sweep and vibrate in joint sand, compact wet, and install edge restraint before handoff — the job isn't done until the restraint is spiked and the surface is swept clean.

Tools & methods

Plate compactor — reversible, 4,000-5,000 lbf centrifugal force for residential jobs, larger (5,500+ lbf) for commercial/vehicular; always with a protective pad on the first post-lay pass to avoid chipping paver faces.
Screed rails and screed board (commonly 3/4" PVC pipe or aluminum rail) for a uniform bedding-sand depth across the bed.
Wet-cut or vacuum-shrouded masonry/paver saw for border and pattern cuts — same silica-control rule as any masonry cutting (OSHA 1926.1153 Table 1).
String lines and story poles for slope and elevation control across the bed, checked against a laser or transit level on larger areas.
Spiked plastic or aluminum edge restraint, spiked every 12" (300mm), 18" on curves.
Nuclear or dynamic cone penetrometer density testing on commercial jobs where the spec calls for documented Proctor compliance; on residential work, a visual/proof-roll check (no visible rutting or pumping under a loaded wheelbarrow) substitutes. Filled take-off tables and the compaction/lift schedule live in references/playbook.md.

Communication style

To the client: leads with what they can see and touch — pattern, color, edge line — but always states the slope direction and base spec in writing before the deposit, because a buried decision they didn't agree to in writing becomes a dispute at year two when a section settles. To a GC or landscape architect: submittal-style, citing the base/bedding spec and drainage grade against the project's stated design (ICPI Tech Spec 4 reference when asked to justify depth). To a crew: pace and sequence first ("border's set, cut list is on the tailgate, field starts left corner"), flags a bad joint-width check immediately rather than letting the crew keep laying past it.

Common failure modes

Judging the base by the finished surface alone — a beautifully laid pattern over an under-compacted or too-thin base looks identical to a correct job until the first heavy freeze-thaw or wheel-load season.
Compacting the bedding sand before laying pavers, which pre-loads it and prevents the final compaction pass from letting units settle into true interlock.
Skipping or under-spiking edge restraint to save time on a job with tight vehicle-turning geometry, then getting called back for a spreading border within a year.
Laying flat or reverse-graded sections because the visible high point looked fine at eye level without a level check.
The overcorrection — after one callback on a thin base, over-specifying an 8-10" base and 80mm pavers on every patio and walkway regardless of load, which prices the job out for no structural reason.

Worked example

Situation. 800 sq ft residential driveway extension, 20 ft wide x 40 ft long, replacing a cracked concrete apron. Soil test (hand auger + local extension office reference) shows heavy clay subgrade, estimated CBR 3%. Client wants a 60mm paver because "that's what the last house had" — a patio-grade spec quoted by a competitor at $9.50/sq ft.

Naive read. Follow the competitor's quote: 60mm paver, 4" compacted base — standard patio spec, no geotextile. On a driveway over CBR 3% clay, ICPI Tech Spec 4's structural design guidance calls for substantially more base under vehicular load on weak subgrade; a 4" base on this subgrade would show visible rutting under normal car and delivery-truck turning loads within one to two wet seasons, well inside the warranty period the quote implies.

Expert reasoning. Vehicular load plus CBR 3% subgrade moves this to an 80mm paver on 8" (200mm) compacted dense-graded aggregate base, with a woven geotextile separation layer against the clay to stop base fines from pumping up into the base voids under repeated wheel load (ICPI Tech Spec 2's guidance for weak/fine-grained subgrades). Slope: fall the full 40 ft length at 1.5% toward the street, away from the garage slab — 0.015 x 40 ft = 0.6 ft = 7.2" of total fall across the run.

Take-off, reconciled:

Base volume: 800 sq ft x (8/12 ft) = 533.3 cu ft / 27 = 19.76 cu yd compacted. At ~1.4 tons/cu yd compacted density: 19.76 x 1.4 = 27.7 tons → order 28 tons dense-graded aggregate.
Bedding sand: 800 sq ft x (1/12 ft) = 66.7 cu ft / 27 = 2.47 cu yd. At ~1.35 tons/cu yd: 2.47 x 1.35 = 3.34 tons → order 3.5 tons ASTM C33 concrete sand.
Pavers: 4"x8" unit at 4.5 units/sq ft coverage = 800 x 4.5 = 3,600 units net. Add 10% for cuts (soldier-course border plus driveway-edge corners) = 360 → 3,960 units. Supplier pallets at 480 units/pallet: 3,960 / 480 = 8.25 → order 9 pallets = 4,320 units (360-unit buffer covers future repairs from a matched lot).
Joint sand: polymeric, 80 sq ft coverage per 50-lb bag at a 3mm joint width on 80mm pavers: 800 / 80 = 10 bags.
Geotextile: 800 sq ft plus 6" overlap allowance per seam, roll width 12.5 ft → 3 strips of ~65 ft, order a 150 ft roll (one roll covers it with the standard commercial roll length, no seam waste beyond the overlap).

Cost delta vs. the naive quote: base tonnage roughly doubles (28 tons vs. ~14 tons for a 4" base) and paver thickness step-up plus geotextile add material cost, moving the job from $9.50/sq ft to approximately $13.75/sq ft — a $3,400 increase on 800 sq ft the client needs to see and approve before excavation starts, not discover as a change order.

Deliverable — spec sheet sent to the client before deposit (quoted):

> Driveway extension — base and material spec, 800 sq ft

> Subgrade: heavy clay, est. CBR 3% — requires vehicular-grade base, not patio spec.

> Base: 8" compacted dense-graded aggregate over woven geotextile separation fabric, compacted in two 4" lifts to 98% Standard Proctor density.

> Bedding: 1" screeded concrete sand (ASTM C33), uncompacted before laying.

> Pavers: 80mm interlocking concrete, herringbone field with soldier-course border, cut units kept ≥1/3 full unit at all edges.

> Drainage: 1.5% slope full run, falling 7.2" over 40 ft toward the street, away from the garage slab.

> Joints: 2-5mm, polymeric sand, compacted wet after sweep-in; edge restraint spiked every 12" on straight runs, 18" on the curved apron.

> Materials: 28 tons base aggregate, 3.5 tons bedding sand, 4,320 pavers (9 pallets), 10 bags polymeric joint sand, 150 ft geotextile roll.

> Revised price: $13.75/sq ft ($11,000) — reflects vehicular base spec vs. the $9.50/sq ft patio-grade quote previously discussed. Approval needed before excavation.

Going deeper

references/playbook.md — load when taking off a job or sequencing installation: base-depth-by-CBR tables, coverage-rate and bag-yield tables, the lift-compaction and joint-sanding sequence.
references/red-flags.md — load when a base, bed, or finished surface doesn't look right: thresholds for when to stop and re-check rather than lay over it.
references/vocabulary.md — load when writing a spec sheet or talking to a GC/architect: terms generalists blur (interlock, confinement, proof-roll) with the misuse each invites.

Sources

Interlocking Concrete Pavement Institute (ICPI, a unit of NCMA), *Tech Spec 2 — Construction Procedures for Interlocking Concrete Pavements* — base/bedding sequencing, compaction, and joint-sanding procedure.
ICPI, *Tech Spec 4 — Structural Design of Interlocking Concrete Pavement for Roads and Parking Lots* — base-thickness-by-subgrade-CBR design guidance, paver thickness by application.
ANSI/ICPI A60.1, *Specification for the Construction of Interlocking Concrete Pavement* — joint-width tolerance, edge restraint requirements.
ASTM C936/C936M, *Standard Specification for Solid Concrete Interlocking Paving Units* — unit dimensional tolerance and thickness classes.
ASTM C33/C33M, *Standard Specification for Concrete Aggregates* — bedding and joint sand gradation.
ASTM D698 (Standard Proctor) / D1557 (Modified Proctor) — compaction density testing methods referenced by ICPI base specs.
David R. Smith (ICPI technical director), *Permeable Interlocking Concrete Pavements* manual (ICPI/NCMA) — subgrade evaluation and geotextile-use guidance, widely cited across conventional and permeable segmental pavement design.
OSHA 29 CFR 1926.1153 (Respirable Crystalline Silica in Construction, Table 1) — wet-cutting/dust-control requirements for paver and masonry saws.
No direct segmental-paver-installer practitioner has reviewed this file yet — flag corrections or gaps via PR.

View SKILL.md source on GitHub · maturity: draft

Jurisdiction: US (baseline)