Flying Probe Testing: Complete Guide

A prototype board or a low-volume run faces a testing problem that high-volume production never sees: no test program exists yet, and building a fixture for a few dozen boards is not economical. Flying probe testing solves that problem by measuring electrical integrity without any fixture, which is why it is the default test method for prototypes, engineering validation, and small-to-medium production batches.

This guide covers how flying probe testing works, what it checks, how it compares with in-circuit test, which test points and files the method needs, and when it is the right choice for your board.

What Is Flying Probe Testing?

Definition

Flying probe testing is an electrical test method in which precision-controlled probes move across a printed circuit board and touch test points directly, following a software program written for that board. The method is fixtureless: it needs no bed-of-nails fixture, because the probes are the contact points and they move to wherever the program directs them. On an assembled board, flying probe testing is also called fixtureless in-circuit testing (FICT), because it performs many of the same component-level checks as an in-circuit test without the fixture.

The method has been in commercial use since 1986 and has become the standard choice for boards that do not justify a custom test fixture. Flying probe testing works on both bare boards and assembled boards. On a bare board it verifies the copper network; on an assembled board it checks connections and component values. The flexible, programmable nature of the test means a design change costs a program update rather than a new fixture. For the bare-board side of the method, see the PCB bare board testing guide.

Why it matters

Flying probe testing matters because it removes the two barriers that make electrical testing expensive for small batches: fixture cost and fixture lead time. A bed-of-nails fixture costs on the order of twenty thousand dollars and takes weeks to build, which is unjustifiable for a prototype or a short production run. Flying probe testing replaces that fixed cost with a variable programming cost that is available the moment the design data is ready.

For design teams, this changes the testing workflow. A prototype can be tested in days instead of weeks, faults can be found and fixed before the design locks, and the same test program adapts when the layout changes. For procurement teams, it means small batches can carry real electrical verification rather than being shipped on trust. In both cases, flying probe testing is the fixtureless bridge between design and reliable production.

How Flying Probe Testing Works

Test program creation

Every flying probe test starts with a test program, and the program starts from the board’s design data. The engineer imports Gerber, ODB++, or CAD files into the flying probe tester, and the system builds a test plan that maps every net, component pad, and test point. ODB++ is the format flying probe systems read most directly, because it carries the netlist, component placement, and layer data in one file. The program defines which points the probes will touch and what measurement each contact must produce.

Program creation is where flying probe testing differs most from fixture-based methods. Building a bed-of-nails fixture takes weeks of design and manufacturing; building a flying probe program takes a fraction of that time and costs nothing in tooling. When the board design changes, the program is updated instead of the fixture being rebuilt. This is the reason flying probe testing adapts so quickly to prototypes and revision cycles, and it is why the method is the practical starting point in the broader PCB testing guide.

Probe movement and contact

Once the program is loaded, the board is placed in the tester and high-precision cameras align and calibrate the moving probes for micron-level accuracy. The probes move on XYZ axes and contact the board’s test points one at a time or in small groups, approaching pads from whatever angle the program requires. Because the probes move freely, they can reach fine-pitch components, microvias, and dense areas that a fixed fixture cannot access.

A flying probe tester carries between two and eight probes depending on the system. Single-head machines use two to four probes and suit prototype work and low volumes. Multi-head machines use four to eight probes that work in parallel, cutting test time for medium-volume runs and complex boards. The probes make light contact with the surface, so boards experience minimal physical stress compared with the pressure a fixture applies, reducing the risk of damage during testing.

Measurements and reporting

At each contact, the tester measures the electrical property the program expects: continuity of a net, isolation between nets, resistance, capacitance, or inductance. The results are compared against the pass-fail limits in the program, and any open trace, wrong component, or polarity error is flagged. The tester compiles the results into a report that shows exactly which nets or components failed and where they are located, which speeds debugging and rework.

The serial nature of the measurement defines the method’s economics. Because the probes move from point to point and measure sequentially, a test cycle takes longer than a fixture-based test that measures many points at once. A cycle that takes thirty seconds on an ICT system can take up to an hour with flying probes. The trade-off is acceptable for small batches, where the fixture saving outweighs the slower cycle, but it is the reason flying probe testing yields to ICT at high volume.

What Flying Probe Testing Checks

Flying probe systems run a layered set of measurements. Standard programs verify the electrical integrity of every net, and advanced programs add component-level and signal-quality checks. The table summarizes the two levels.

Level	What it verifies
Standard tests	Opens, shorts, connectivity across nets, component presence, resistance, capacitance, inductance (RLC)
Advanced tests	Micro-shorts, high-resistance defects, diode and IC input impedance, phase difference, high-voltage stress

Standard tests

The standard flying probe program verifies opens, shorts, and continuity across every net on the board. An open is a break in a trace that leaves a net disconnected; a short is an unintended connection between two nets. Both faults are the most common manufacturing defects on bare and assembled boards, and both are exactly what a continuity check is designed to find. The test measures each net against the netlist and flags any path that is broken or bridged.

Beyond connectivity, standard programs check component presence and value. The tester measures resistance, capacitance, and inductance (RLC) at component locations and compares the readings with the expected values from the design. This catches missing parts, wrong-value parts, and components placed in the wrong orientation where polarity matters. Cameras on most systems automatically inspect component polarity at the same time, adding visual verification to the electrical measurements. For bare boards, acceptance criteria follow the IPC-A-600M standard, which defines what a conforming bare board looks like.

Advanced tests

Advanced flying probe programs extend coverage into fault classes that standard continuity checks miss. Micro-short detection finds tiny bridges between adjacent fine-pitch pads that pass a normal resistance check. High-resistance defect testing detects joints that are physically connected but carrying degraded current, a common precursor to field failure. Diode and IC input impedance measurements verify semiconductor junctions and signal integrity at critical nodes.

Phase difference measurement and high-voltage stress testing round out the advanced set. Phase difference analysis compares signal timing between points to catch routing or component faults that pure resistance checks cannot see. High-voltage stress, also called hi-pot testing, verifies insulation between high-voltage areas on the board. These advanced features give a modern flying probe tester coverage that approaches fixture-based in-circuit test, which is why many manufacturers use it as a fixtureless ICT replacement. Qualification and performance of the finished board follow the IPC-6012 standard, and the measurement methods trace to the in-circuit testing (ICT) guide covers the fixture-based side of the decision.

When to Use Flying Probe Testing

Flying probe testing is the right choice when fixture cost or lead time would block testing, and the scenarios below are the situations where it delivers the most value.

• Prototype validation. A new design needs testing before it is trusted, and a prototype rarely justifies a fixture. Flying probe tests prototypes quickly without tooling, so faults are found while the design can still change.
• Early design and engineering validation. During R&D, the layout changes frequently. The test program updates in minutes, so every revision is tested with the same equipment and the same measurement baseline.
• Low-volume and medium-volume production. Batches that do not justify an ICT fixture carry full electrical verification with flying probe, which is the most cost-effective way to test them.
• High-density and fine-pitch boards. Flying probes reach fine-pitch components, microvias, and dense areas that a fixed fixture cannot access, making the method ideal for compact and advanced boards.
• Custom and irregular board shapes. Non-rectangular and unusual form factors fit the flexible, non-fixture process without the cost of a custom fixture tool.
• Time-sensitive projects. When the schedule is tight, the lack of fixture lead time is decisive. Flying probe programming is ready in days, not the weeks a fixture requires.

Flying probe testing also serves as the starting point in a staged test strategy. A new design starts on flying probe, and if it matures to high volume, the production line adds in-circuit test once the volume justifies the fixture. Running both gives each stage the fastest and most economical coverage, which is why flying probe remains the flexible foundation of most test plans.

Test Points for Flying Probe

Flying probe testing depends on test points, and the placement and size of those points decide how much of the board the probes can reach. A board designed without test points can only be partially tested, because the probes have nowhere safe to land. Designing for flying probe means placing accessible points on every net that needs verification.

Size and spacing

Flying probe systems tolerate smaller test points and looser spacing than ICT fixtures, because the probes move to each point individually and can be aimed with micron-level accuracy. A test point as small as about 0.5 millimeters in diameter typically works for flying probe, and points on a grid are not required the way they are for a bed-of-nails fixture. This makes flying probe the forgiving option for boards where space is tight and test points are difficult to place.

Even with that tolerance, larger test points give more reliable contact. A bigger pad gives the probe a larger landing area, reducing missed contacts and false failures, and it also withstands repeated probing without wearing through. The practical rule is to use the largest test point the layout allows, keep the pad finish solderable, and record every point in the test file so the program can target it precisely. The PCB test points design guide gives the sizing tables and clearance rules.

Placement rules

Placement determines whether a test point is reachable. Keep test points clear of components taller than the probe clearance, because a probe cannot touch a pad hidden under a connector or a tall component. Place points on the outer layers wherever possible, and avoid covering them with solder mask. Dedicated test points are always preferable to probing component pads, which risks damaging the component and produces unreliable readings.

Distribute test points across the board rather than clustering them, so the probes have room to travel and the program can test nets in parallel on multi-head systems. On double-sided boards, plan which side each point sits on so the tester can probe both sides without obstruction. For pads and probe access detail, the test pads and SMT test points article covers finishes, probe marks, and the design rules that keep coverage high and damage low.

Test Files & Program Preparation

A flying probe test program is built from your board’s design data, so the files you provide determine how quickly and accurately the program is created. The most important file is ODB++, which flying probe systems read directly because it packages the netlist, component placement, test point locations, and layer definitions in a single format. Providing ODB++ removes a conversion step and the errors that step can introduce.

If ODB++ is not available, the test program can be built from the Gerber files plus a separate netlist and a component centroid file. The netlist defines every net and the components and pins attached to it, and it is what the continuity and RLC checks verify against. The centroid file gives placement coordinates for every part. Confirm the exact format your contract manufacturer accepts before transferring files, and review the generated test program against the netlist on the first build to catch data-format mismatches early.

Test point information completes the package. The program must know where every test point sits, its size, and which side of the board it is on. Provide this data alongside the netlist so the probes target the correct locations. A complete, accurate data package is the fastest route to a working flying probe program, and it is the practical requirement behind the PCB testing guide principle that test files decide test accuracy.

Flying Probe Testing Cost Considerations

Flying probe testing changes the cost structure of electrical testing. Instead of paying a fixed fixture cost that only makes sense at volume, you pay a per-board cost driven by test time. For prototypes and small batches, that structure is dramatically cheaper than an ICT fixture that costs tens of thousands of dollars for a run of a few hundred boards. The economics are why flying probe is the default for low-volume electrical verification.

Three factors set the per-board cost of a flying probe test. Test time is the largest: the more test points and measurements, the longer the cycle, and the higher the per-board cost. Board complexity follows, because dense boards take more probes to cover. Volume determines how the fixed programming cost spreads, since a program written once is used across every board in the run. For small batches the programming cost dominates; for larger batches the per-board cycle dominates.

Flying probe testing also carries hidden savings beyond the fixture. There is no fixture to store, maintain, or rebuild when the design changes, and no fixture lead time that delays production starts. Boards experience less mechanical stress than in a fixture press, reducing damage and rework. The flying probe tester cost guide compares equipment purchase and outsourced testing in detail, and our PCB testing services covers flying probe as part of a full test flow.

Flying Probe Testing Checklist

Use this checklist to prepare a board for flying probe testing and confirm coverage before you commit to production.

• Test points placed on every net that needs verification
• Test point size and finish match the probe requirements
• ODB++ file (or Gerber + netlist + centroid) provided
• Test point coordinates confirmed for both board sides
• Test program reviewed against the netlist on first build
• Board size and component height within the tester limits
• Pass-fail limits defined for RLC and continuity measurements

FAQ

What is a flying probe test?

A flying probe test is an electrical test method in which programmable probes move across a PCB and touch test points directly to measure continuity, shorts, resistance, capacitance, and component values. It needs no test fixture, which makes it ideal for prototypes, low volumes, and frequently changing designs.

How does a flying probe tester work?

The tester imports the board’s design data, builds a test program that maps every net and test point, then moves probes on XYZ axes to touch each point and measure the expected electrical values. Results are compared with pass-fail limits and compiled into a report that flags open traces, shorts, and wrong components.

What is the difference between flying probe and ICT?

Flying probe testing is fixtureless and moves probes to each point serially, so it has no fixture cost but a slower cycle. In-circuit test uses a custom bed-of-nails fixture that contacts every point at once, so it is faster but requires a fixture costing tens of thousands of dollars. Flying probe suits low volumes; ICT suits high volumes.

Can flying probe testing be used on bare boards?

Yes. Flying probe systems test both bare boards and assembled boards. On a bare board, the tester verifies trace continuity and isolation across the copper network before components are placed. The same method and many of the same programs apply to both stages.

What test files does flying probe testing need?

The preferred file is ODB++, which packages the netlist, component placement, test point locations, and layers in one format. If ODB++ is not available, provide Gerber files, a separate netlist, and a component centroid file. Test point coordinates for both sides of the board complete the package.

How long does a flying probe test take?

It depends on the number of test points and the board’s complexity. Because the probes measure serially, a cycle can run from minutes to roughly an hour for a fully populated board. Multi-head systems with four to eight probes cut that time significantly by testing points in parallel.

Does flying probe testing damage the board?

Flying probe testing applies light contact and no fixture pressure, so boards experience minimal mechanical stress compared with a bed-of-nails press. Using correctly sized test points and durable pad finishes further reduces any risk of probe marks or wear.