Why Humidity Alone Cannot Predict PCB Condensation Failure

A PCB passes its humidity test. Later, the assembled product develops intermittent leakage during a cold start.

Before changing the laminate or increasing conductor spacing, ask a more basic question:

Did the qualification test reproduce the surface conditions that occurred in service?

Relative humidity is useful, but it does not describe the complete electrical environment on a PCB. Board temperature, surface residues and applied voltage can change how moisture affects the circuit.

For products exposed to temperature transitions, condensation deserves its own place in the reliability review.

Humidity describes the air. Condensation happens on the surface.

Relative humidity depends on air temperature. It does not, by itself, tell you whether a particular region of the board is wet.

A board entering warm, humid air may remain colder than its surroundings. If its surface falls below the local dew point, condensation can form.

This matters when reviewing operating scenarios such as:

• Cold equipment starting in a warmer environment.
• Electronics exposed to enclosure temperature changes.
• Assemblies located near cooled surfaces.
• Products repeatedly moved between storage and operating conditions.

These are scenarios to assess, not evidence that every product will experience condensation.

The practical question is whether the actual board surface can reach conditions that the qualification profile did not reproduce.

Leakage current and ECM are different observations

Moisture can reduce surface insulation resistance and increase leakage between conductors.

Electrochemical migration, or ECM, involves an additional process: metal dissolves, ionic species move through an electrolyte under electrical bias, and metal deposits can grow into conductive filaments.

A leakage increase therefore does not, on its own, prove that a dendrite has formed. Electrical monitoring and physical examination answer different parts of the failure investigation. Published PCB research describes the electrolyte formation, metal dissolution, migration and deposition stages of ECM. [1]

This distinction matters during troubleshooting. A board may malfunction because insulation resistance has dropped before a permanent metallic bridge is identified.

Why the shape of the water layer matters

Research by Rao and colleagues examined transient condensation using in-situ imaging and electrical measurements. The reported test conditions included surface undercooling values of 1.2°C and 8.2°C, with applied voltages of 2 V, 10 V and 48 V.

The researchers identified droplet coalescence and the development of a continuous water path between conductors as important steps in accelerating electrochemical failure under their test conditions. [2]

The engineering implication is straightforward: the presence of droplets alone does not fully describe the risk. How water connects across the surface also matters.

Those experimental voltages and temperature differences are not universal acceptance limits. They should not be copied into a qualification plan without considering the product geometry, surface condition and operating environment.

Surface residues belong in the review

A board’s moisture response also depends on what remains on its surface.

Research on transient climatic conditions has shown that hygroscopic flux residues can promote local water-layer formation and corrosion-related reliability problems. [3]

For an assembled product, the review should therefore include the actual process sequence:

• Soldering and flux selection.
• Cleaning, where applicable.
• Handling and storage.
• Rework.
• Coating preparation and application.

A successful result from a clean test coupon may not describe an assembly with different residues or inaccessible areas.

The useful question is not simply, “Was the board cleaned?” It is, “Does the evidence represent the surface condition of the product being qualified?”

What a passed humidity test tells you

A passed test provides evidence for the tested combination of temperature, humidity, duration, electrical loading, specimen preparation and acceptance criteria.

The next step is to compare that combination with the intended service conditions.

NPL describes condensation testing that controls board cooling independently of chamber humidity and uses surface insulation resistance measurements to monitor electrochemical effects. This illustrates why chamber humidity and board-surface condensation need to be considered separately. [4]

For products with a credible condensation exposure, the following review can help define the remaining questions.

This is a planning framework. The test conditions and limits should follow the product’s requirements and risk assessment.

Build the evidence around the failure mechanism

When investigating a suspected condensation problem, start by recording the sequence of events.

What were the air temperature and humidity? What was the board temperature? When was power applied? Did leakage begin during cooling, warming or drying?

Then connect the electrical evidence to the physical evidence.

A useful investigation should distinguish among:

• Reduced insulation resistance.
• Metallic migration.
• Corrosion or loss of conductor continuity.
• Other moisture-related functional changes.

This prevents a visible residue or stain from becoming the assumed root cause before the electrical failure mechanism has been established.

Questions to settle before qualification

Before releasing a PCB assembly intended for a condensation-prone environment, align the design, manufacturing and reliability teams on five points:

1. The expected environmental transitions.
2. The board’s powered and unpowered states during those transitions.
3. The surface condition represented by the test specimen.
4. The electrical and functional monitoring required.
5. The acceptance criteria and failure-analysis plan.

These decisions turn a general request for “humidity resistance” into a testable engineering requirement.

References

[1] Nanoscale monitoring of the initial stage of water condensation on a printed circuit board. Heliyon, 2025.
https://doi.org/10.1016/j.heliyon.2025.e42117

[2] Rao, J. M., Mantis, I., Lakkaraju, A. R., and Ambat, R. Effect of condensed water layer morphology in triggering leak current and electrochemical migration failure in electronics. Surfaces and Interfaces, 98, 110355, 2026.
https://doi.org/10.1016/j.surfin.2026.110355

[3] Residue-Assisted Water Layer Build-Up under Transient Climatic Conditions and Failure Occurrences in Electronics. DTU Research Database.
https://orbit.dtu.dk/en/publications/residue-assisted-water-layer-build-up-under-transient-climatic-co/

[4] National Physical Laboratory. Condensation testing of electronic assemblies.
https://www.npl.co.uk/products-services/electronics-reliability/condensation-testing-of-electronic-assemblies