Bare PCB Fabrication: What You Get Before Assembly

What a Bare Board Is

A bare board, also called a plain circuit board, is a printed circuit board that has finished the fabrication stage and has not entered assembly. Everything the circuit needs to be built on is present: the laminate, the copper pattern, the drilled holes, the solder mask, the legend, and the surface finish. Nothing is populated. There are no resistors, no capacitors, no connectors, and no chips.

The distinction sounds trivial until a project mixes the two stages up. Fabrication and assembly are separate processes, usually quoted separately, and a bare board is the deliverable that transfers between them. Designers who treat the handoff explicitly get fewer surprises than those who assume the board house and the assembly house are looking at the same drawing.

Bare Board and Assembled Board

The assembled board, usually written PCBA, contains the same base board plus the components and the solder joints that attach them. Its function is complete; the bare board’s function is to make that completion possible.

In practice the difference shows up in what has to be specified. A bare board order is defined by the stackup, the copper weight, the hole schedule, the finish, and the fabrication tolerances. An assembly order is defined by the bill of materials, the paste and stencil strategy, the reflow profile, and the inspection plan. Keeping those two lists separate makes it much easier to place the work, and much easier to see which supplier is responsible when something is wrong.

What the Board Contains

The substrate provides mechanical support and electrical insulation. It is the layer that determines thickness, stiffness, and thermal behaviour, and it is chosen first.

The copper is patterned into conductors, and its weight determines how much current the traces can carry and how finely they can be etched. Heavier copper supports higher current and finer copper supports finer features, and the two goals pull against each other on the same layer.

The solder mask covers the copper that must not be soldered, protecting it from oxidation and preventing bridges between adjacent pads. The legend marks reference designators for assembly and inspection. The surface finish, whether hot air solder leveling, ENIG, immersion silver, immersion tin, or OSP, protects the exposed pads and determines how long the board remains solderable in storage.

Materials

FR-4 is the default for the great majority of bare boards: glass-reinforced epoxy that is mechanically strong, electrically stable, and available in a wide range of thicknesses and Tg grades. High-Tg versions are chosen when a board will see repeated reflow cycles or must stay dimensionally stable at temperature.

Aluminium and other metal-core laminates are used where heat has to leave a power device quickly, which is why they dominate LED lighting and many power modules. PTFE and other low-loss materials serve radio-frequency designs where the dielectric loss of FR-4 is too high. Polyimide serves flexible and rigid-flex boards, and ceramic serves package substrates and applications where a very low expansion coefficient is required.

Board Types

Single-sided boards carry one copper layer and are the least expensive construction. Double-sided boards route on both faces and connect through plated holes. Multilayer boards stack four, six, eight, or more layers with blind, buried, or through vias, and they are the norm once a design needs a reference plane, controlled impedance, or significant routing density.

Rigid boards are the standard. Flexible boards use a thin polyimide core and bend in service. Rigid-flex boards combine rigid sections for components with flexible sections for interconnection, which is common in products that must fold into a tight enclosure.

How a Bare Board Is Built

The work starts from the design files: Gerber data for each layer, a drill file, and a stackup drawing. A mistake in any of them is expensive later, so the first fabrication step in a good shop is a review of the data against the process capability.

Inner layers are imaged with dry film resist, exposed, developed, and etched, then inspected. The layers are laminated under heat and pressure, with the resin flowing to fill the copper topography and bond the stack. Drilling creates the holes, and the walls are cleaned before copper is plated through them so that the layers connect electrically.

Outer layers are then patterned, the solder mask and legend are printed, and the surface finish is deposited. Electrical testing checks for opens and shorts on every net, and final inspection confirms the appearance and the dimensions. Each of those steps has tolerances, and the ones that matter most are the registration between layers, the hole position, the dielectric thickness, and the finish thickness.

Specifying a Bare Board

The stackup should be a drawing rather than a table in an email, because it defines the dielectric heights that control impedance and the overall thickness that fits the enclosure. Where impedance matters, the target value and tolerance belong on the same drawing.

Decide the surface finish before the assembly process is fixed. OSP is inexpensive but has a limited shelf life and is sensitive to handling, while ENIG is flat, durable, and suited to fine-pitch parts and long storage. Also state the fabrication class and the inspection level, since the same board built to a standard class and to a high-reliability class differs in acceptance criteria and in price.

Finally, decide how the boards will be delivered. Panels and arrays suit automated assembly, while individual boards and breakaway rails suit prototyping, and the choice affects both the pricing and the handling.

Testing and Quality

Electrical test is the baseline. A flying probe or a bed-of-nails fixture checks continuity and isolation across the whole net list, which catches the defects that visual inspection cannot. Cleanliness, solder mask adhesion, and hole quality are checked by inspection and, on critical products, by cross-sectioning.

Thermal stress testing, ionic contamination testing, and microsection review are added when the product’s reliability class calls for them. On high-reliability products these checks are part of the specification rather than optional extras, and a supplier with a documented quality management system can usually explain exactly what was measured and what the acceptance limits were.

Where Bare Boards Are Used

Almost every electronic product passes through a bare board stage. Consumer devices buy bare boards and populate them, industrial controllers and automotive modules do the same on thicker, higher-reliability laminates, and medical and communications equipment adds the testing that the product class demands.

Engineers also buy bare boards on their own for development: a single-sided or two-layer prototype can be turned around in days, which lets a design be tested before the tooling and assembly schedule are committed. Where the design is still changing, that is usually the cheapest way to learn.

What Drives the Cost

Layer count is the largest single factor, because every layer adds imaging, lamination, and test steps. Board size, copper weight, hole count, and surface finish follow. Small quantities carry a higher unit price because the setup is spread over very few pieces, which is why panel-level orders look dramatically cheaper per board.

Two habits keep bare board cost under control: keeping the design inside standard capabilities so no special tooling, tighter tolerances, or extra inspection is required, and ordering the quantity that the schedule actually needs rather than the quantity that looks tidy on the order form. A fabricator able to quote both PCB manufacturing and the PCB assembly that follows will frequently spot a design rule that would have added cost at one stage and caused a defect at the next.

FAQ

Is a bare board the same as a substrate? In everyday use, yes. Both describe a fabricated board with no components, although substrate is also used for the base material alone and for package carriers.

Does a bare board work on its own? No. It has no active or passive components, so it cannot perform a function until it is assembled.

What do I need to order one? Gerber data for every layer, a drill file, a stackup drawing, the solder mask colour, the surface finish, the copper weight, the quantity, and the fabrication class.

Why test a bare board electrically? Because opens and shorts inside a multilayer stack cannot be seen, and finding them after assembly is far more expensive than finding them at the factory.

How should bare boards be stored? In clean, dry packaging with the finish protected. OSP has the shortest shelf life, and handling with bare hands degrades any finish.

Conclusion

A bare board is the physical result of everything decided at the design stage: the stackup, the copper, the holes, the mask, and the finish. Keeping those decisions explicit, and keeping them separate from the component decisions that follow, is what makes the transition into assembly predictable. For the steps on either side, see our notes on PCB design and layout and SMT PCB assembly.