Inspection engineering · RAYPCB
Flying-probe testing for PCB and PCBA
Evaluate flying probe around the board state, accessible features, required tests and build quantity. Compare setup and cycle costs only when the competing test scopes are equivalent.
Share the product revision, required checks and expected records. Early questions are welcome.
When is flying probe a useful choice?
It can reduce dedicated fixture work for a changing design or lower-volume build, but cycle time, access and coverage still matter. Bare-board continuity testing and assembled-board measurements are different programs; the method name alone does not describe the scope.
How does a flying-probe test work?
A flying-probe test system moves probes to programmed contact locations and makes electrical measurements through those contacts. The measurement sequence and available instruments depend on the tester and the product. Moving probes replace some of the fixed-contact tooling used in a dedicated fixture; they do not remove the need for a controlled test program or a physical support strategy.
The same broad contact principle can be used for unpopulated boards and assembled boards, but the electrical tasks differ. A bare-board program checks intended networks for continuity and isolation. An assembled-board program can make selected component or network measurements where access, circuit topology and tester resources permit them. Do not assume a bare-board result verifies the populated product.
State the board condition at quotation. A page that simply promises “flying-probe testing” leaves important questions unanswered: before or after assembly, which networks, which component checks, powered or unpowered operations, what exclusions and what evidence? Defining those details produces a meaningful scope instead of a machine-name comparison.
Bare board versus assembled-board coverage
| Board state | Typical task | Critical qualification |
|---|---|---|
| Bare PCB | Continuity and isolation of the released conductive networks | Match the design netlist, production data and test limits |
| Populated PCBA | Selected component values, networks or structural checks | Parallel paths, inaccessible nodes and tester options affect testability |
| Powered or combined test | Specific powered checks where the system and fixture support them | Requires an approved power, stimulus, load and safety plan |
| Repair investigation | Targeted measurements around an observed fault | Program findings do not automatically establish root cause |
Seica’s bare-board tester overview is a primary equipment reference for the bare-board use of moving probes. Available functions differ between systems. Treat vendor feature lists as prompts for a capability discussion, not evidence that all features are supplied on every RAYPCB order.
For fabrication requirements, continue to PCB electrical testing. For an assembly with recurring production volume, compare the access and measurement needs with ICT planning. Functional verification still requires a separately defined product test specification.
Design test access before the first build
Identify the nets and features that must be measured, then verify that suitable contact surfaces are reachable on the delivered board or panel. Probe approach, neighboring component height, surface condition and support can constrain access. A test point that appears exposed in the CAD top view may be blocked by a tall part or an actual fixture boundary.
Agree the contact geometry with the intended tester. Avoid copying a generic minimum pad diameter or component clearance from an unrelated machine specification. The permissible landing area depends on probe type, positioning capability, board variation and the required contact reliability. Protect sensitive surfaces or high-reliability connections when probe marks could matter.
Multiple pads on one electrical net can improve access, but they do not automatically add independent fault coverage. Conversely, one accessible node may not allow a component to be isolated from surrounding parallel paths. The electrical schematic and test strategy are as important as the physical coordinates.
For two-sided assemblies, define handling, support and the order of operations. The program must recognize the actual variant and DNP population. Releasing the wrong coordinates or a different board origin can create setup problems even if the netlist is otherwise correct.
Files that make the quotation and program review useful
Provide the applicable CAD test data, netlist, BOM, placement information, schematic and assembly drawing as agreed with the test team. Identify the product revision, variants, critical measurements and test-point map. Some tasks can be prepared from less data, but missing information can reduce coverage or increase engineering effort.
Specify the expected test report and traceability. If you need serial-level results, measured values or an explicit list of untested items, make those requirements visible before the program is built. A generic pass/fail summary may not support an investigation that depends on a particular node or measurement.
Use a review loop for exceptions. The test engineer can identify inaccessible points, circuit interactions or missing requirements. The product owner then decides whether to change the design, accept a documented limitation or add another method. Avoid treating an untestable feature as passed simply because the overall program completed.
A useful program-release review also identifies the representative units used for verification, the expected responses and the treatment of known anomalies. Keep these references controlled when a design changes. Verification against one accepted unit alone does not demonstrate every defect-detection capability; review the relevant fault objectives and representative evidence with the test engineer.
Flying probe versus ICT: calculate the cost crossover
A flexible setup can be attractive when designs change or quantities are limited. A dedicated fixture can be economical for a stable repeat build when its lower running cost offsets its setup. There is no universal quantity at which one method becomes cheaper. Compare the same required coverage and include handling, test time, engineering and fixture work consistently.
Flying Probe and Fixture Budget Comparison
The model uses total cost = setup cost + quantity × variable cost per board. It reports both totals and a crossover quantity where one exists. Enter costs from comparable supplier quotations or your internal planning data, all in the same currency. The tool contains no RAYPCB price list or current market rates.
Variable cost can include test time, operator handling and agreed per-board reporting if the quotes use that basis. Keep common fabrication or assembly costs outside both sides unless they differ for a real reason. Retests, maintenance, freight and future revisions are excluded unless you deliberately incorporate them into the entered values. The example is purely illustrative.
Cycle time, repeatability and future revisions
The number of probe moves is only one part of cycle time. Measurements may require settling, switching, guarding, charge or discharge steps and handling. A shorter cycle is useful only when it preserves the required scope and the measurement is repeatable. Ask the supplier to distinguish machine test time from the total production cycle.
On a repeat design, fixture condition, contact wear and program changes can alter the effective economics. On a changing design, reprogramming may be simpler than replacing a dedicated interface, but it still requires verification. Include anticipated revisions and product mix in the business decision instead of relying on the first batch alone.
Keep apparent contact failures separate from confirmed product defects. A controlled retest procedure can help diagnose an interface problem. Unrestricted retries until a pass appears can conceal intermittent faults. Record the sequence and disposition where the agreed traceability requires it.
Specify standards, limits and complementary inspection
For bare boards, the contractual electrical-test specification defines connectivity requirements; IPC-9252B’s public scope is a useful reference, with its maintenance status and construction limits considered explicitly. For assembled boards, use the applicable assembly acceptance requirements and the product-specific electrical test plan. A flying-probe machine does not confer IPC compliance by itself.
Electrical access to a net is different from seeing a hidden solder interface. Use X-ray inspection for selected hidden-geometry questions and AOI for programmed visible features. A passing functional test may still leave structural defects outside the observed operating conditions, so keep the coverage limitations visible.
For a prototype, request a testability review early enough to change the layout if necessary. For repeat manufacturing, provide the expected lot sizes and design stability. Return to PCB Assembly for populated-board work or PCB Manufacturing for bare boards. The test-method comparison helps coordinate the complete route.
Questions behind a flying-probe quotation
Use these fields to scope the report with your quotation. The table is a planning example, not a RAYPCB test result or a promise that every record is included by default.
| Record item | Information to include | Why it matters |
|---|---|---|
| Object | Bare board or assembly, revision, quantity and accessible sides | The assembled population changes access and measurement possibilities. |
| Coverage | Required nets or measurements, access map and exclusions | Count accessible required items, not probe moves. |
| Economics | Program/setup, fixture alternative, cycle, handling and retest basis | Compare equal scopes and customer-supplied costs. |
| Change | Program revalidation and revision/fixture change allowances | A changing design can alter the economical choice. |
Walk through an engineering example
Illustrative planning: a fixture has a larger setup allowance but a smaller per-unit test allowance. Their crossover is useful only if both plans test the same required characteristics. If one omits a required powered measurement, the cheaper total does not answer the purchasing decision. Use the calculator with actual quotations, not assumed market rates.
Continue to review fixture-based ICT access before comparing quotations.
The email link opens a draft for you to complete. Contact is a separate form; these pages do not send messages automatically.
Questions engineers and buyers ask
Is flying probe only for prototypes?
No. Its suitability depends on program scope, throughput, access, product mix and economics. Prototypes are a common use, but a fixed quantity cutoff is misleading.
Can it test components under a BGA?
Accessible nets may support selected electrical checks, depending on the circuit and tester. Direct physical access to every hidden joint is not implied. Review the coverage and complementary methods.
Does fixtureless mean no tooling charge?
No. Programming, board supports, alignment, engineering and handling can still require setup work. Compare the quotation’s actual included scope.
Why does the calculator show no crossover?
If setup and variable costs do not trade off, one option may remain cheaper at every positive quantity, or the two models may be equal. The tool reports those cases without inventing a crossover.
Turn the inspection question into an agreed scope
Send the current files, product revision, quantity and the requirements that matter to your application. Identify any unfinished criteria so the engineering review can address them explicitly.
Request the method, coverage, acceptance basis and report format with your quotation. Tool briefs can support the discussion; the Contact form does not automatically receive this page’s entered data.
Engineering references reviewed September 2026. Illustrations explain methods; equipment photographs do not establish product-specific coverage, inspection results or certification.