A fabricated PCB is an empty stage. It has laminate, copper traces, drilled holes, plated barrels, and a surface finish, but on its own it does nothing. What turns that bare board into a functioning product is the population of components — resistors, capacitors, inductors, diodes, transistors, connectors, and integrated circuits — placed, soldered, and verified during assembly. Fabrication decides where a part can sit and how reliably it can be joined; assembly decides whether it actually gets there. Understanding that handoff is what separates a design that passes first-article inspection from one that stalls in rework.
Component Categories That Shape the Assembly Plan
The word “component” covers a very wide range of physical objects, and each category imposes its own demands on the assembly floor. At the small end sit passive surface-mount parts, including chip components down to 01005s, which are placed by machine because no human hand is steady enough. At the other end sit large processors in ball-grid packages — PBGA, CBGA, TBGA, FPGA, CGA, and LGA — where hundreds or thousands of solder joints sit underneath the body and cannot be inspected visually. Fine-pitch micro BGAs at 0.4mm pitch push placement accuracy further still, and connectors, shields, and power devices often arrive as through-hole parts that must be soldered from the opposite side of the board.
Between those extremes are the parts that make assembly interesting: QFN packages with exposed thermal pads, fine-pitch leaded ICs, sockets, headers, and the occasional odd-form part that does not fit a standard feeder. Boards that combine surface-mount and through-hole parts are described as mixed technology, and they require a deliberate sequence rather than a single pass.
The board itself sets hard boundaries on what can be assembled. A fabrication capability that supports BGA pitch down to 0.15mm and minimum trace width and spacing of 0.002″ is only useful if the assembly stage can place and join parts at that scale. Via-in-pad structures — filled with conductive paste, non-conductive material, or a copper plug — matter here too, because an unfilled via under a BGA pad wicks solder away from the joint and creates voids. Heavy copper, up to 30 oz., holds heat and changes how joints form on power sections.
Surface finish is the other half of the compatibility question. ENIG, ENEPIG, hard gold, gold fingers, OSP, immersion silver, immersion tin, and wire-bondable gold all present different surfaces to molten solder. A gold-finger edge connector needs hard gold for wear resistance; a board that will be wire-bonded needs a finish suited to that process; an OSP coating protects copper during storage but demands attention to shelf life before assembly begins. Choosing components without reference to the finish and pad geometry is one of the most common reasons a design that looks correct on screen fails on the line.
Assembly: Placing, Soldering, and Verifying Components
Assembly is the stage where the bill of materials becomes physical. Surface-mount technology handles the majority of modern parts, through-hole soldering handles connectors and legacy components, and mixed-technology builds combine the two on a single board. Production equipment supports board sizes up to 20″ x 24″, which covers large backplanes and panelized arrays as well as compact modules.
The work does not end when the last joint solidifies. Verification is part of assembly, not an optional extra. X-ray inspection looks through package bodies to check BGAs and QFNs, where solder joints are hidden beneath the part, and it also catches problems on bare circuit boards before any part is placed. IC programming loads firmware into microcontrollers and memory devices so the board leaves the line ready to run rather than waiting for a separate programming step. Functional testing then confirms the populated board behaves as a circuit rather than merely looking correct under magnification.
Because different board types behave differently, assembly has to be organized around the substrate: rigid PCB assembly, flex PCB assembly, and rigid-flex PCB assembly each bring their own handling concerns. A thin polyimide flex circuit cannot be treated like a 2.4mm rigid board, and a rigid-flex build with a cavity needs fixtures that protect the transition zones.
For teams moving from prototype to production, the practical advantage of keeping fabrication and assembly under one roof is continuity. When the same operation drills the microvias, applies the finish, and then places the pcb board components, questions about pad size, via fill, or finish compatibility get answered inside the same workflow instead of bouncing between vendors. Short-run and prototype assembly benefit most, because the feedback loop between design intent and physical result is shortest.
When Board Technology Changes the Component Plan
The more advanced the board, the more the assembly plan must adapt. On an HDI design, microvias filled with epoxy and copper-capped, blind and buried vias drilled mechanically or by laser, staggered, stacked, or skipped microvia stacks described as 9+N+9, and cavity structures all change the surface a part lands on. A cavity board may place a component inside the board thickness rather than on top of it, which alters handling, fixturing, and inspection entirely. Back drilling, with a minimum hole size of 15.7 mils and depth tolerance of ±6 mils, removes unused via stubs to protect signal integrity — but it also means the assembly team must understand which holes are intentional and which are artifacts.
Radio-frequency boards raise the stakes further. Materials from Rogers, Isola, Arlon, Taconic, and Panasonic MEGTRON families are selected for dielectric constant, loss tangent, and thermal expansion behaviour, and the copper surface roughness influences skin effect as signals travel. Impedance control held to ±5% is only meaningful if the assembled board preserves it, which puts pressure on joint quality, pad geometry, and any added capacitance from via-in-pad fill. Hybrid stackups, where high-frequency material is combined with standard FR-4 in the same board, are common in microwave and antenna products and require the assembly plan to account for two different thermal behaviours on one panel.
Rigid-flex and flex builds introduce mechanical concerns that rigid boards never face. Features down to 25µm and flexible dielectric cores down to 25µm mean the board can bend, and every soldered joint in a bend region is a potential fatigue point. Eccobond epoxy strain relief is used to anchor joints where flexing occurs. Substrates such as polyimide (Kapton), PET, PEN, and LCP each respond differently to heat, so a profile that works on one flex material may distort another. Sculptured flex circuits, castellated holes, press-fit pins, countersunk holes, and semi-flex constructions each add a step that must be planned before the first part is placed.
A realistic example: a compact antenna module using low-loss laminate, a fine-pitch BGA, and a rigid-flex tail to a connector. The fabrication side supplies the hybrid stackup, the laser-drilled microvias, and the ENEPIG finish; the assembly side must place the BGA without voiding under the pads, program the device, X-ray the hidden joints, and confirm that the flex tail still meets its bend requirements after soldering. Every one of those outcomes traces back to a decision made about components and board together, long before the first panel entered the line.
Lagos architect drafted into Dubai’s 3-D-printed-villa scene. Gabriel covers parametric design, desert gardening, and Afrobeat production tips. He hosts rooftop chess tournaments and records field notes on an analog tape deck for nostalgia.