SHT30 Humidity Drift Above 80% RH: Condensation, Recovery and FHT30 Qualification

What does “SHT30 drift above 80% RH” mean?

Engineers often use drift for any result that differs from a reference. That shorthand hides several mechanisms. Calibration error is the equilibrium difference between a device and a traceable reference. Hysteresis depends on whether the sensor arrived from a dry or humid condition. Long-term drift develops over months or years. Prolonged high humidity can add a temporary offset as the sensing polymer retains a memory of recent conditions.

Condensation, contamination and test temperature error are different again. A temporary high-RH offset may fade after the device returns to its normal range. A thermal-gradient error disappears when the fixture reaches equilibrium. Liquid water requires dry-out and inspection of the full assembly. Contamination from cleaners, adhesives or outgassing plastics may recover only partly—or not at all.

Why 80% RH is a qualification boundary, not a cliff

The SHT3x-DIS datasheet example—about +3%RH after 60 hours above 80%RH—identifies direction, scale and exposure time. It is not a guaranteed shift for every unit or environment. Actual behavior depends on RH level, temperature, dwell time, prior history, airflow, contamination and lot variation. A brief excursion to 82%RH is not equivalent to days near saturation.

It is also easy to mistake test error for sensor offset. Relative humidity is strongly temperature dependent. Sensirion's testing guidance says the reference and device under test must be at the same absolute temperature. Heat from the PCB, cables, chamber fan or nearby electronics can keep one sensor warmer than another and create an apparent RH disagreement.

High humidity and condensation require different tests

Air at 90%RH can still be free of liquid water. Condensation begins when the sensor or nearby surface drops below the local dew-point temperature. It can occur during cold start, defrost, rapid temperature ramps, outdoor night cooling or movement from an air-conditioned room into humid air.

Once droplets form, an RH value near 100% is no longer a complete description. Water can bridge conductors, trap residues, promote corrosion, block a sensing opening or cause a long dry-out. A chamber wall covered with droplets does not prove that the sensor is wet, and a saturated RH reading does not prove it either. Record surface temperature, calculated dew point and visible wetting separately.

How should an SHT30 high-humidity drift test be diagnosed?

1. Establish a controlled baseline. Stabilize the DUT and a calibrated reference together at a moderate condition. Record signed error across several devices and lots.
2. Control temperature before judging RH. Place the reference close to the DUT, minimize gradients and log both temperature channels alongside RH.
3. Preserve the exposure history. Record ramp rate, dwell, RH, temperature, airflow, measurement cadence and heater activity.
4. Watch the digital path. Log raw words, CRC failures, bus errors, resets and status flags.
5. Return to baseline and keep logging. A reversible offset should trend toward its initial value. Persistent error points to contamination, damage, corrosion, reference trouble or a thermal effect.
6. Inspect materials and process history. Board wash, solvent vapor, uncured adhesive, coating and packaging outgassing can affect polymer humidity sensors.

What recovery procedure applies to SHT30?

Sensirion describes passive recovery after an SHT30 returns to its normal temperature and humidity range. For severe exposure, its June 2025 SHTxx handling instructions suggest an accelerated reconditioning sequence: bake at 100–105°C and below 5%RH for 10 hours, then rehydrate at 20–30°C and about 75%RH for 12 hours.

That is a Sensirion SHTxx procedure. It is not automatically safe for a populated assembly, battery, connector, adhesive, display or enclosure, and it must not be copied to FHT30 without NYFEA's controlled instruction. Save the as-exposed and passively recovered data before any accelerated treatment.

Both devices publish an integrated heater function. Heater operation changes the local sensor temperature and therefore the reported RH. It does not replace enclosure drainage, vent design, contamination control or a defined recovery protocol.

Where FHT30 fits

NYFEA FHT30 is a candidate for evaluation in SHT30-class digital humidity-sensor designs. Its published page lists a 2.5 mm × 2.5 mm DFN-8 package, 2.0–5.5 V supply, I²C operation up to 1 MHz, selectable 0x44 and 0x45 addresses, CRC-protected measurement words, an integrated heater, a 0–100%RH range and typical accuracy of ±3%RH.

Those similarities define a shortlist. They do not establish pin compatibility, command compatibility, timing equivalence, equal accuracy, identical high-humidity behavior or equal production qualification.

Choose FHT30 for evaluation when the design team can repeat the SHT30 baseline and revalidate command timing, error budget, enclosure behavior and production conditions. If those conditions cannot be controlled, the published feature match is not enough to support a replacement decision.

A practical FHT30 high-humidity qualification plan

This is an engineering template, not a manufacturer specification. Set pass limits from the application error budget before testing and use enough samples and lots to expose variation.

1. Baseline: measure signed RH and temperature error, noise, CRC rate and reset behavior on the production PCB.
2. Non-condensing high RH: step above 80%RH while keeping every surface above dew point; include the real worst-case dwell.
3. Humidity cycling: cycle between moderate and high RH to quantify hysteresis and accumulated offset.
4. Condensation challenge: if the product can cross dew point, define the wetting mechanism and monitor faults, recovery and corrosion risk separately.
5. Passive recovery: return to baseline and log recovery time, residual offset and samples that fail to converge.
6. Final-product confirmation: repeat the worst case with the production vent, membrane, coating, adhesive and nearby heat sources.