Flying probe and bed-of-nails testing can both check whether a fabricated multilayer PCB has the intended electrical connections. Flying probe usually fits changing designs and smaller batches because it avoids a dedicated contact fixture. Bed-of-nails testing can reduce recurring test time when a stable design and sufficient repeat volume justify the fixture investment. Neither method is automatically more thorough: coverage depends on the reference netlist, accessible nodes, measurement settings, and test program.
For engineers and purchasing teams, that distinction matters before expensive components reach assembly. A low testing price offers little protection if the program checks the wrong revision or excludes an inaccessible branch. This guide compares bare-board electrical testing methods, explains the multilayer constraints behind coverage, and identifies what to put in an RFQ so competing quotations describe equivalent work.
Bare-board electrical testing checks conductive networks before components are fitted. Continuity measurements look for missing or excessively resistive connections within an intended net. Isolation measurements look for unintended connections or unacceptable leakage between nets that should remain separate.
The distinction is particularly useful for a multilayer PCB, where internal routing and via connections are not fully visible after lamination. A board can look acceptable externally while containing an electrical discontinuity inside its interconnect structure.
The public IPC-9252B scope and purpose excerpt explains that electrical testing checks connectivity against design requirements, while other checks are needed for characteristics such as dimensions and solder mask. Passing electrical test does not establish that a board satisfies every manufacturing or assembly requirement. ipc.org
Keep the production stages separate. Bare-board testing cannot confirm the values, placement, or solder joints of components fitted later. A PCB assembly test plan may therefore also require inspection, in-circuit testing, and functional testing, each with its own acceptance criteria.
A flying probe tester moves probe heads to accessible conductive features and performs programmed measurements. It avoids a dedicated pin fixture, although board supports, data preparation, programming, and verification still require work.
This flexibility helps during engineering changes: the supplier can update the program rather than necessarily rebuilding a contact fixture. However, a new revision still needs access checks and program validation. “Fixtureless” does not mean “no engineering cost” or “instant release.”
A bed-of-nails arrangement uses a board-specific contact fixture or adapter to connect many test locations to the analyzer. Spring-loaded probes or other contact arrangements align with the board’s accessible features.
Once contacts are established, the analyzer can switch measurements without repeatedly moving probe heads across the board. That can improve throughput. Contacting many nodes together does not mean every net-to-net measurement happens simultaneously; measurement sequence and analyzer architecture still matter.
Fixture suitability also depends on alignment, board support, contact force, and maintenance. A worn or contaminated contact can cause a false failure and trigger unnecessary troubleshooting.
The following comparison describes typical engineering trade-offs, not guaranteed machine specifications or universal quantity thresholds.
| Decision factor | Flying probe | Bed-of-nails / fixture testing |
|---|---|---|
| Initial preparation | Program development and validation; no dedicated nail fixture | Program development plus fixture or adapter engineering and validation |
| Recurring test time | Depends on travel, accessible points, measurements, and handling | Often shorter once contacts are established; still depends on the test sequence |
| Design revisions | Program changes are generally easier to accommodate | Contact-location changes may require fixture modification or replacement |
| Dense features | Can suit small accessible targets when the machine supports them | Feasibility depends on fixture technology, pin geometry, and support |
| Electrical coverage | Determined by access, source data, and programmed checks | Determined by access, source data, and programmed checks |
| Production fit | Often attractive for prototypes, mixed products, and revision changes | Often attractive for stable, repeat production |
| Commercial risk | Recurring machine time can become a bottleneck | Fixture investment may become stranded after an early design change |
| Evidence to request | Coverage review, limits, validated cycle time, and exceptions | The same evidence, plus fixture revision and maintenance controls |
Avoid treating flying probe as exclusively a prototype method. Mycronic’s PCB and substrate test systems overview describes flying probe use for larger quantities and automated systems for medium production volumes. The practical decision therefore depends on the supplier’s actual equipment and workload, rather than a fixed industry-wide crossover quantity. mycronic.com
Neither method guarantees better coverage simply because of its name. Ask what the supplier measures, which connections are reachable, and how exceptions are controlled.
First, separate boards tested, connections checked, and isolation combinations evaluated. Testing every board does not, by itself, prove that every intended connection and required isolation relationship was assessed. “100% electrical test” needs an agreed denominator.
A net with several branches requires more thought than confirming that one pair of points conducts. A measurement between two accessible points may miss an open on a different branch. Ask how the program represents endpoints, branches, and excluded locations.
Similarly, one accessible contact on a net may support an isolation measurement against another net, but it cannot establish continuity through every branch of that first net. A coverage report should distinguish these cases.
Request the isolation strategy, including how the program identifies relevant neighboring nets. On multilayer boards, relationships across layers can matter as well as spacing on a single surface. The appropriate selection method depends on the agreed test requirements and the supplier’s validated process.
Also clarify whether the program uses direct resistance measurements, indirect signature comparisons, or a combination. These are measurement strategies, not synonyms for flying probe or fixture testing. Do not assume a machine category tells you which strategy was used.
An application concerned with small resistance increases may need a more discriminating measurement than a basic open/short screen. Hioki’s FA1283 technical documentation describes four-terminal low-resistance measurements for detecting resistance increases associated with via defects. HIOKI
That documents an equipment capability. Buyers still need confirmation that the supplier’s equipment, probe access, program, and limits support the required measurement. It also does not turn an electrical test into a prediction of service life.
Layer count alone is a poor basis for choosing a test method. The more useful questions concern net topology, connection access, board mechanics, and the measurements needed.
Internal connections need an accessible electrical path. Probes do not physically contact buried vias inside a finished board. A buried interconnect can be evaluated electrically through accessible nodes connected to it, where the topology and measurements permit. A fully inaccessible internal network cannot be assumed covered by final surface probing.
Pad geometry and surface condition affect contact. Small lands, recessed features, solder-mask openings, and the surface finish can influence repeatability and probe marks. Ask the supplier to assess the actual PCB layout against its equipment and contact rules. Avoid applying one generic pad diameter to every tester.
Mechanical support affects the result. Thin boards, large panels, bow, twist, and accumulated fixture force can complicate stable contact. Flying probe also requires reliable positioning and support. Request the proposed holding arrangement rather than assuming either method is mechanically neutral.
Resolve access during PCB design, especially if production may later move to a fixture. An early prototype that passes flying probe does not establish that the same contact geometry will support economical fixture testing.
A test program can accurately verify the wrong reference. If a connectivity error already exists in the manufacturing data and the test netlist is extracted from the same data, a finished board may match that error and still pass.
Altium’s explanation of PCB netlist checking distinguishes a netlist generated from Gerber data from an independently exported IPC-D-356 netlist used to compare design connectivity with manufacturing outputs. That comparison can identify discrepancies before fabrication begins. Altium
For purchasing, this creates two separate deliverables: a documented data comparison before manufacture, and electrical testing of physical boards against the approved reference. Request both when the project warrants them.
Maintain consistency between the released PCB schematic, layout, manufacturing files, and test data. Record the part number and revision, and resolve intended net ties or approved CAM changes before releasing the program. An independent netlist still cannot establish that the underlying circuit design is correct.
When a supplier finds a mismatch, define who approves the resolution. Quietly changing the reference to make the discrepancy disappear removes the value of independent checking.
Compare total cost over the expected quantity of the same usable revision, rather than only the first order or an optimistic lifetime forecast.
A simple starting model is:
Flying probe total = flying probe setup + quantity × flying probe recurring cost
Fixture total = fixture setup + quantity × fixture recurring cost
If fixture testing has a lower recurring cost, the simplified crossover is:
Crossover quantity = additional fixture setup cost ÷ recurring saving per board
For an illustrative calculation only, suppose flying probe setup is $200 with $1.60 recurring cost per board, while fixture setup is $1,400 with $0.40 recurring cost. The crossover is 1,000 boards. These invented inputs demonstrate the calculation; they are not supplier quotations or market benchmarks.
The real comparison needs additional terms: fixture modification, maintenance, storage, spare contacts, expected retesting, and program revalidation. Confirm whether quotation quantities refer to individual boards or panels. Comparing a panel-level test charge with a board-level charge produces a misleading result.
Design stability can outweigh forecast volume. A fixture justified by anticipated annual demand may lose its value if contact locations change before repeat orders arrive. Conversely, a stable repeat design can make fixture investment worthwhile even when each individual release is modest.
Schedule also has a cost. Ask for measured cycle time on the intended program, loading and unloading time, available capacity, and fixture completion timing. A fast analyzer cannot recover time lost waiting for a fixture or an overloaded test department.
A passing report describes the board under the specified electrical test conditions. It does not establish controlled impedance, laminate integrity, copper geometry, or resistance to future thermal and mechanical stress.
For example, an interconnect may conduct during final electrical test without the test establishing its durability through the product’s operating life. Additional inspection, coupons, microsection evaluation, or reliability testing may be appropriate where the drawing and application require them. These checks answer different questions and should have separate acceptance criteria.
Review the supplier’s quality control requirements as a coordinated plan. Ask which checks apply to every board, which use samples or coupons, and which require a separate order. Do not infer coverage from a list of available inspection machines.
Retest rules deserve attention too. Repeatedly probing a failing location until it passes can obscure an intermittent contact or board defect. Require a defined investigation and disposition process, with traceable final results and retention of failure history where agreed.
Provide enough information for the supplier to propose a test method against a defined acceptance scope. Ask for a coverage review before accepting a price comparison.
| RFQ item | Information to provide or request | Why it affects the decision |
|---|---|---|
| Released data | Part number, revision, manufacturing outputs, drill data, and an independent netlist where available | Establishes the authorized reference and detects data mismatches |
| Board construction | Stack-up, via structures, dimensions, thickness, finish, and panel format | Supports access, support, handling, and program assessment |
| Quantity and stability | First batch, expected repeats, and likely engineering changes | Determines fixture amortization and revision exposure |
| Electrical acceptance | Contractual standard and revision, applicable test level, and approved measurement limits | Makes supplier quotations technically comparable |
| Coverage | Boards to be tested, continuity scope, isolation strategy, and documented exclusions | Prevents ambiguous “100% tested” statements |
| Special checks | Low-resistance measurement, impedance, dielectric withstand, or other application-specific tests where required | Identifies work beyond the ordinary open/short program |
| Delivery evidence | Report contents, lot or unit traceability, program revision, and retest records | Makes acceptance and later investigation practical |
| Commercial terms | Setup charges, recurring charges, fixture ownership, storage, modification, and replacement | Exposes costs beyond the initial test price |
Ask the supplier to return three concrete items: its recommended method, the proposed coverage and exceptions, and an itemized price and schedule. Approve the test scope before authorizing tooling. This makes the choice between flying probe and bed-of-nails a documented engineering and purchasing decision.
Not inherently. Measurement accuracy depends on the instrument, contact quality, measurement configuration, and calibration. Coverage is a separate issue: either method can leave connections unchecked if access or programming is incomplete. Compare the actual measurement requirements and coverage report.
It can detect some electrical faults involving buried vias through accessible nodes connected to those vias. It cannot physically probe the buried structure. Detection depends on topology, accessible endpoints, and measurement sensitivity; a pass does not prove long-term via reliability.
No. Testing every unit describes board population coverage. It does not define continuity coverage, the isolation relationships evaluated, or excluded features. Require those details separately, including the reference netlist and the acceptance limits used.
There is no universal crossover quantity. Calculate it from actual setup and recurring charges, then consider design changes, fixture life, handling, and available capacity. Use the expected quantity of the same revision rather than total product-family demand.
No. Bare-board electrical testing assesses the unpopulated interconnect. Assembly testing addresses components and their connections, while functional testing evaluates specified powered behavior. A fabrication test cannot validate component placement, firmware, or final product performance.
Provide released manufacturing data, drill information, stack-up and fabrication requirements, revision identification, and an independent electrical netlist where available. Add quantities, acceptance requirements, special measurements, and reporting expectations. Confirm which data formats the supplier can process before release.
Lorem ipsum dolor sit amet consectetur adipiscing elit dolor
Learn how PCB impedance coupons, TDR measurements, tolerance limits, and traceable reports support fabrication acceptance and more accurate supplier RFQs.
Compare flying probe and bed-of-nails for bare multilayer PCBs, including electrical coverage, netlist control, probe access, fixture economics, and RFQ requirements.
Learn the differences between PCB surface preparation, wet processing, and final finishes, with practical guidance on plating terminology, inspection evidence, and RFQ requirements.
