Quick answer
ICT uses a board-specific fixture to check accessible nets and components efficiently on a stable design. Flying probe uses programmed moving probes where flexible access and lower fixture commitment matter. Functional test powers the assembly to verify defined behaviors, interfaces and measurements. Choose the combination only after defining failure risks, test access, build stage, quantity and the evidence required for acceptance.
ICT, flying probe and functional test do not ask the same question. In-circuit test and flying probe can localize selected structural or component-level faults through accessible electrical nodes. Functional test powers the assembly and evaluates behaviors, interfaces or measurements defined for the product. The useful plan starts with the required evidence, not with the name of a tester.
This distinction matters because a populated PCBA test plan is different from an unpopulated board continuity test. IPC-9252B addresses electrical test for unpopulated printed boards; it does not define a complete PCBA acceptance plan. For an assembled product, connect the method to the released design, available access, build stage and the specific records needed for acceptance.
What evidence should release the PCBA?
Before selecting equipment, write the release question in observable terms. Examples include: identify opens or shorts on accessible nets, verify selected component values or orientations, confirm that a power rail starts safely, exercise a communication interface, or record a project-defined output under a stated load. Each question needs an owner, setup, limit and failure disposition.
PCBArise publishes PCB assembly testing as a project-specific capability. That does not mean ICT, flying probe or functional test is included by default. The quotation and approved test plan must identify the method, access, inputs, limits, unit or lot scope and evidence required for the actual build.

What in-circuit test verifies
ICT uses a board-specific fixture, test program and accessible test points to contact many nodes in a repeatable setup. Depending on the implemented program and access, it can check selected opens, shorts, component values, orientation-related signatures and other structural conditions. Its strength is fault localization on a stable design when the required nodes are designed for fixture access.
ICT does not automatically prove complete end-use behavior. Inaccessible nodes, components outside the program and behavior not stimulated by the setup remain outside its evidence. Fixture design, revision stability, access geometry, program limits and maintenance therefore belong in the release decision—not only the label “ICT.”
What flying probe verifies
A flying-probe system moves electrical probes to programmed targets instead of relying on a full bed-of-nails fixture. It can exercise selected nets and components that are reachable by the probes and supported by the program. That flexibility can be valuable during NPI, for variants or where a dedicated fixture commitment is not yet justified.
Moving probes do not remove access constraints. Target size, keep-outs, component height, board support, probe travel and the executable test library still shape the plan. Flying probe also does not prove every end-use behavior or every feature that cannot be reached. Quantity and cycle-time context should be reviewed for the actual job rather than reduced to a universal threshold.
What functional test verifies
Functional test powers the PCBA or product-level assembly and observes defined behavior. The setup may use connectors, loads, simulated inputs, firmware, instruments or project-specific fixtures to evaluate rails, interfaces, outputs, controls, sensors or other agreed functions. Its value is evidence about the behavior actually exercised under the stated conditions.
Functional test is only as complete as its specification. A board can pass the exercised sequence while an untested function or latent assembly fault remains. Safe power-up, firmware and configuration, I/O mapping, loads, limits, measurement uncertainty, result format and failure disposition all need to be controlled. It should not be described as a universal substitute for structural localization.
ICT vs flying probe vs functional test comparison
| Method | Primary evidence | Access and setup | Planning fit | Does not prove alone |
|---|---|---|---|---|
| ICT | Accessible nets, selected components and structural faults defined by the program | Board-specific fixture, planned test points, program and limits | Stable revision and repeat fixture execution | Complete end-use behavior or inaccessible features |
| Flying probe | Selected reachable nets and components through programmed probe moves | Access map, probe program, keep-outs, support and limits | NPI, variants or changing quantity/revision context | Unreached features or complete end-use behavior |
| Functional test | Defined powered behavior, interfaces and measurements | Safe power, connectors or fixture, firmware, loads, sequence and limits | Any stage needing agreed product-behavior evidence | Every assembly fault or any behavior not exercised |
The methods can be complementary. A structural method may localize a production fault efficiently, while functional test demonstrates that specified behavior works. The final combination depends on risk, access and acceptance requirements, not on a generic ranking.
Five inputs that determine the test route
1. Required evidence and failure risks
List the release-critical failure modes and what observable result would close each one. Separate bare-board continuity, assembly structure, programming and powered behavior. The broader PCB inspection methods guide explains why visual, X-ray and electrical methods also answer different questions.
2. Physical and electrical test access
Review reachable pads, test-point geometry, connectors, ground references, component height, keep-outs and fixture support before the layout is frozen. A DFM review can identify manufacturing questions, but the design and test authorities decide which access and acceptance requirements belong in the product release.
3. Build stage and revision stability
An early prototype may prioritize fast program changes and diagnosis. A stable repeat build may justify a dedicated fixture and controlled baseline. These are planning tendencies, not fixed rules. Revalidate the route after a layout, alternate component, firmware, fixture or acceptance change.
4. Quantity and operational economics
Compare fixture engineering, program development, execution time, maintenance, operator interaction and failure-analysis effort for the actual quantity scenario. Do not choose from quantity alone; a lower setup commitment may trade against execution time, while a fixture investment may be inappropriate before the design is stable.
5. Acceptance records and disposition
Define whether the project needs a unit result, measured values, failure codes, serialized records, lot evidence or only a controlled pass/fail record. Name who reviews failures and who can approve deviation, retest or repair. PCBArise's quality-control content provides manufacturing context, while the project documents define the actual retained evidence.
PCBA electrical test selection flow
The flow can produce one method or a complementary set. If a pilot result leaves a release-critical question unanswered, adjust the access, method, program, limits or product requirement before approving the baseline. A passing result should always be interpretable against the revision and conditions evaluated.
Printable electrical-test requirements checklist
Product and evidence
- □ Board/assembly part number, revision, active variant and build stage are identified.
- □ Release-critical failure modes and the evidence required for each are listed.
- □ Bare-board, assembly-structure and powered-behavior questions are separated.
Access and setup
- □ Test points, connectors, references, keep-outs, support and fixture constraints are reviewed.
- □ Safe power, loads, firmware/configuration, I/O and sequencing are defined where functional test is required.
- □ The selected ICT or flying-probe program scope is tied to accessible nodes and components.
Limits and records
- □ Conditions, units, limits and measurement or comparison method are unambiguous.
- □ Unit/lot scope, result format, serialization and retained evidence are stated where required.
- □ Failure, retest, repair and deviation decisions have named authorities.
- □ Revision, alternate, fixture, firmware or acceptance changes trigger revalidation.
Use the Print / PDF button above to bring this checklist into a test-plan review. It is a planning aid; the approved product requirements, quotation and project test specification remain controlling.
Prepare the test inputs before requesting a route
Send the current assembly data, quantity and build stage, available test access, firmware or programming inputs, required behaviors, conditions, limits and evidence expectations. PCBArise can then discuss a project-specific assembly test route instead of treating the word “test” as a complete requirement. Use the project review page only when the release package is ready; no form submission is required to use this checklist.
Engineering review
Prepared and reviewed for project use
Engineering Director
Senior Quality Engineer
Review scope: technical accuracy, evidence wording, standards references, internal links and release readiness. Project requirements remain subject to the released files, applicable acceptance criteria and agreed test documentation.
FAQ
Questions engineers ask before release
What is the main difference between ICT and functional test?+
ICT primarily checks accessible nets, components and structural conditions through a board-specific fixture and program. Functional test powers the assembly and verifies defined behaviors, interfaces or measurements. They answer different questions and may be complementary.
When is flying probe considered instead of ICT?+
Flying probe may suit NPI, variants or a changing revision or quantity context because it uses programmed moving probes rather than a full bed-of-nails fixture. The choice still depends on reachable targets, execution needs, program scope and the required evidence.
Can functional test replace ICT or flying probe?+
Not automatically. Functional test proves only the behavior exercised under its defined conditions, while ICT or flying probe may localize structural faults on accessible nodes. Select the combination from the product risks and acceptance questions.
Does passing a PCBA test prove zero defects?+
No. A passing result supports only the checks, access, conditions and limits implemented by the approved plan. Unexercised behavior, inaccessible features and latent conditions can remain outside that evidence.
What should be supplied for a PCBA electrical test review?+
Provide the controlled assembly revision and variant, quantity and build stage, test access, firmware or programming inputs, safe power and I/O conditions, required behaviors, limits, result records and the authority for failure disposition.
Reference points
Sources and verification starting points
External standards and industry references help frame the decision. Confirm current supplier evidence and project-specific requirements before release.

