Bare Board Test: Finding Opens and Shorts Before Assembly

A bare board is tested before any component is placed on it, and the purpose is to catch the faults that fabrication can produce: a broken trace, a short between two nets, a missing connection to an inner layer. Finding them at this stage costs a board; finding them after assembly costs the board and the components.

The test is electrical, and it is applied to the finished board rather than to a coupon. What it can find depends on how the board was designed for test, which makes it a layout requirement as well as a manufacturing one.

What the Test Looks For

An open circuit is a connection that should exist and does not: a trace broken by a scratch, a via that was not plated, a barrel that cracked during processing. A short circuit is a connection that should not exist, usually caused by an etching defect or by conductive debris.

Both are detected by measuring the resistance between points that should be connected and between points that should be isolated. The measurements are simple; the challenge is reaching every net and doing it fast enough.

Fixture Based Testing

A fixture with spring pins contacts every test point at once and the machine measures the network in a fraction of a minute. It is the fastest method and the one used for volume production.

The fixture has to be built for the board, and it has to be able to contact the pads without damaging them. A pin that presses too hard marks the pad, which matters if the board is to be soldered afterwards, and a pin that presses too lightly gives an intermittent reading that is worse than a clear failure.

Flying Probe Testing of Bare Boards

A flying probe machine visits the test points in turn and needs no fixture. It is slower, and it is the practical choice for prototypes, for small batches and for a board whose design is still changing.

The machine can also be more tolerant of the pad geometry, since the probes can approach at an angle. For a board with a high density of test points, however, the time per board becomes significant, and the choice between the two methods is essentially the same calculation as the one used for assembled boards.

Test Points and Layout Requirements

A net can only be tested if it has a point that the probe can reach. That means a pad of adequate size, clearance from tall features and a position that is accessible from the side the machine approaches, on a grid that the fixture can support.

Where the design provides test points on all nets, the test is straightforward. Where it does not, the test relies on the pads of components that will be placed later, which works for a fixture but is less reliable for a probe. Designing the test access with the layout is much cheaper than adding it afterwards.

Test Coverage and Its Gaps

Coverage is measured as the proportion of nets that are accessible. A net without a test point is a net whose faults will be found later, if at all, and the gap should be documented rather than discovered.

Some faults are difficult to detect even with full access. A high resistance connection, such as a partially cracked barrel, may pass a continuity test and fail in the field. Where the risk is significant, a more sensitive measurement is used, sometimes at a higher voltage, to make the defect visible.

Electrical Test and Other Inspections

Electrical test finds opens and shorts and says nothing about the geometry of the copper. A trace that is too thin, an annular ring that is marginal or a solder mask defect will pass the test and appear later as a reliability problem.

That is why the electrical test is combined with the visual and the dimensional checks that a fabricator performs. The combination is what makes the finished board’s quality verifiable rather than assumed.

Records and Failure Analysis

Every failure found by the test should be recorded with its location and its type. The pattern of failures is the diagnosis: shorts concentrated in one area point at an etching or a plating problem, while opens along one direction point at handling damage.

The data is also the basis for the discussion with the fabricator. A supplier who receives a list of failures with positions can investigate them; one who receives a batch of boards with no information can only replace them. The same principle applies to fabrication notes and to any other feedback that improves the process.

Test Voltage and Its Selection

The voltage used for the continuity and isolation measurements affects what the test can find. A higher voltage makes a marginal connection visible, and it also risks damaging a board that cannot tolerate it.

The choice is a compromise that depends on the product. For a board with fine features and thin dielectric layers, the voltage has to be limited; for a robust board with wide spacing, a higher value gives more confidence. The limit belongs with the manufacturing tolerances that the board was designed to.

Handling and Storage After the Test

A board that has been tested is still a bare board and is exposed to the same risks as any other. The probe marks left by the fixture are a particular concern, because a marked pad can oxidise faster than the surrounding surface.

Where the board will be stored for a while before assembly, the packaging and the storage conditions matter, and the shelf life applies from the date of fabrication. Handling the tested boards as carefully as untested ones avoids introducing a defect at the last stage of the fabrication process.

Test Data and Its Format

The test produces data as well as a verdict. The resistance values, the location of the failures and the time taken for each board are all recorded by the machine, and they are useful for the fabricator and for the assembly house.

Keeping that data in a form that can be compared across batches is what turns a series of pass and fail results into a trend. It costs nothing beyond a decision to keep it, and it is the same principle that supports the wider quality control programme.

Process Control and Verification

On a design of this kind, short circuit is the item that decides how the rest of the board is arranged. Keeping a sample from the panel turns a dispute into a measurement, because the same coupon can be re-examined by both parties without rebuilding the batch. Reviewing the design before the data is released is cheaper than correcting it after the panel is in the tank, because every step downstream inherits the decision made at the front end.

FAQ

Is bare board test always required? Not always, but it is cheap insurance on a dense board with inner layer connections that cannot be inspected visually.

Can the test damage a board? It can mark a pad if the pin pressure is too high. The fixture should be set up to contact reliably without leaving a mark that affects soldering.

Does a flying probe test take longer? Yes, and the difference is large. It is justified where the volume does not support a fixture.