PCB Inspection After Fabrication: What to Check

Once a board has been designed and built, the difference between a working assembly and a batch of scrap is measured at inspection. The checks are unglamorous, but each one maps to a specific failure mode that would otherwise be discovered by the customer. This is what a PCB inspection should cover after design and fabrication, and why each test exists.

Visual Inspection First

Visual checks are cheap and catch a surprising share of defects, so they come first.

The board should be clean and free of handling damage. Scratches, dents and stains are not cosmetic problems: a scratch deep enough to breach a trace creates an intermittent open, and contamination left on the surface interferes with soldering and can degrade insulation over time under bias and humidity.

Pads and vias come next. Pad shape and size must match the design, with no missing or deformed pads; vias must be clear and unobstructed, with no lifted or flaked plating. A damaged pad prevents correct solder joint formation, and a blocked or poorly plated via affects both signal transmission and current-carrying capacity, which is why annular ring integrity is one of the first things to check on fine-pitch boards.

Then the traces. Conductors should be continuous and clearly defined, with no opens, no shorts and no indistinct edges. Width and spacing must meet the design rules, particularly in dense regions, and impedance-controlled lines deserve close attention: where a trace’s geometry is blurred or reduced, its characteristic impedance changes, and the effect appears as a signal integrity problem that no firmware can fix.

Solder mask and legend are part of the same pass. Mask must cover what it should and stay off what it must not, since a mask sliver on a pad prevents wetting and mask encroachment on a test point makes probing unreliable.

Electrical Testing

The electrical checks verify what the eye cannot see.

Continuity test. Every conductor, pad and via is checked for the connection the design intends, using a multimeter in resistance mode or a dedicated continuity tester. Power and ground nets and any critical signal paths get priority, because a break there stops the assembly immediately. Where an open is found, it has to be repaired before electrical testing continues meaningfully — a board with an open net cannot be judged on its other measurements.

The same pass detects shorts, which is the more dangerous failure: a short between two nets may not show as a functional fault on the bench, but it will surface under load or temperature.

Insulation resistance. Adjacent traces, adjacent pads, and separate power and ground layers are measured to confirm that the resistance between them meets specification. This is the check that detects leakage paths, and it matters most in humid or high-voltage applications, where a marginal value drifts lower in service and the board becomes unstable.

Hi-pot test. A defined voltage is applied across the isolation barrier and the board is watched for breakdown, flashover or abnormal heating. The hi-pot test verifies dielectric strength, which is the property that keeps a mains-referenced circuit isolated from anything a user can touch. It is also the test most likely to destroy an already-marginal board, so it is performed after visual and low-voltage checks rather than before.

How these checks are structured depends on the design intent, and the review that precedes them is described in this PCB layout quality checklist. For boards that carry mains or isolated interfaces, the sequence of low-voltage and high-voltage tests should be written into the test plan rather than decided on the bench.

<img src="https://www.gopcba.com/wp-content/uploads/2026/09/34-2.jpg" alt="continuity and insulation resistance testing of a bare board” />

Reliability Checks

Functional tests under comfortable conditions prove very little about a board that will live in a hostile environment.

Environmental exposure. Boards are placed in high temperature, low temperature and humid conditions, and their behaviour is compared against the specification. Under heat, joint cracking and copper lifting show up; under humidity, insulation resistance falls and leakage paths appear. The usual pattern is that a board passes at the extremes and fails during or after a transition, which is why the profile matters as much as the endpoints.

Mechanical stability. Shock and vibration are applied to simulate the service environment, and the assembly is inspected for loosened or detached components and for cracked solder joints. On products that move or vibrate continuously, this test is where the mechanical design of the assembly is actually verified — pad geometry, underfill, adhesive and stiffening all show their value or their absence here.

What Each Check Actually Catches

It helps to keep the failure modes in view, because inspection is only useful when it targets something.

Visual inspection catches process escapes: plating damage, contamination, mask faults and handling damage. Continuity testing catches layout and fabrication errors: broken traces, missing connections, blocked vias and unintended bridges. Insulation resistance and hi-pot testing catch material and spacing problems, including the ones that only become dangerous in service. Environmental and mechanical testing catch the design decisions that a functional test cannot see, such as a stackup that does not tolerate thermal cycling or a package that cannot survive vibration.

None of these overlap much. A board can pass continuity testing and still fail hi-pot; it can pass hi-pot and still crack in thermal cycling. That is the argument for running the full sequence rather than the cheapest subset.

Where the Checks Belong in the Flow

Inspection is not a single gate at the end, and it is not a substitute for design verification, which has to happen while the board is still data. A board that reaches fabrication with an unresolved doubt will carry that doubt through every test that follows. Electrical testing belongs after fabrication and before assembly for bare boards, and again after assembly at the level the product requires. Reliability testing belongs after assembly, on production-representative units, because the failure modes it exposes combine the board with the solder joints and components.

Design inputs determine most of the outcome here, since a specification that names the acceptance criteria is what allows a test to produce a pass or fail rather than an opinion. The information a manufacturer needs for this, from the schematic to the test requirement, is set out in this summary of PCB design inputs and steps.

Finally, the checks are only meaningful when the shop can build to the tolerances being checked. A design that pushes trace width, spacing or via size close to the process limit will generate marginal inspection results indefinitely, and the durable fix is to review those choices against real capability, as described in these notes on evaluating a PCB design company.

FAQ

Is visual inspection enough for a new design? No. It catches process defects, not electrical ones. A board can look perfect and still have a blocked via or a bridge that only appears under load.

When should the hi-pot test be run? After visual inspection and low-voltage electrical testing, and before functional test on production units. Running it first risks destroying boards that would otherwise have been correctable, which wastes both the panel and the information.

How long should environmental testing last? Long enough to cross the transitions, not just to sit at the extremes. The failures that matter in service usually occur during a change of condition rather than during a soak.