Pressure sensor drift in metal housing: what to check

Pressure sensor drift in metal housing needs a controlled diagnosis, not an immediate change of alloy. Deposits and galvanic corrosion are plausible when stainless steel shares water with aluminum or zinc. However, mounting stress, temperature and reference-pressure changes can produce similar symptoms. Start by checking the actual pressure, then compare sensor condition, mounting and electrical connections. A dry electronics compartment does not rule out problems at the wetted diaphragm or reference vent.

What the drift pattern points to

Record the direction, magnitude and timing of drift before removing or cleaning anything. These patterns help separate possible causes:

• Drift appears only in a metal housing: investigate electrical coupling, housing corrosion and differences in mounting stiffness or heat transfer.
• A white film develops on the diaphragm: investigate mineral scale and corrosion products, without assuming the film identifies its source.
• The reading changes after tightening or inserting the sensor: investigate mounting stress, seal compression and bore alignment.
• The error follows water or housing temperature: investigate thermal compensation, differential expansion and temperature-dependent mechanical loading.
• The reading changes with weather: check whether an absolute sensor is responding to atmospheric pressure above the water.
• The sensor recovers slowly after drying: investigate reference-vent moisture, seal relaxation and reversible surface effects. Recovery alone does not identify the mechanism.
• Only one sensor drifts: compare its depth, orientation, deposits, bore fit, wiring and reference path against the unaffected position.

A polymer housing comparison is useful, but it changes more than electrical conductivity. It can also change stiffness, thermal behavior and seal loading.

Which causes to investigate first

For a wetted mixed-metal assembly with visible deposits, investigate surface contamination and electrical coupling early. They are credible leading suspects, not a confirmed diagnosis.

Use this practical checking order:

1. Confirm actual pressure, temperature and the sensor's pressure-reference arrangement.
2. Inspect the diaphragm, pressure opening and surrounding housing for deposits or corrosion.
3. Check mounting stress, O-ring condition and any trapped cavities.
4. Map electrical connections between the sensor, housing, cable shield and supply.
5. Investigate sensor damage or internal instability with the sensor manufacturer if the earlier checks remain inconclusive.

Galvanic corrosion needs an electrolyte, an electrochemical driving force and an electronic connection between the metals. Water provides the ionic path; it does not replace the electronic connection.

Aluminum or zinc can corrode preferentially when electrically coupled to stainless steel in suitable water chemistry. The outcome depends on surface condition, water conductivity, oxygen, geometry and exposed area. The aluminum versus stainless steel comparison covers broader material selection; it does not establish the cause of a sensor offset.

Establish a pressure and temperature baseline

Identify whether the sensor measures absolute, vented gauge, sealed gauge or differential pressure. These arrangements do not respond identically to changes in the surrounding atmosphere.

A vented gauge sensor needs its documented reference path to remain dry and unobstructed. A submerged cable joint or blocked vent can change the reference pressure even when the electronics remain dry. A sealed reference cavity can also respond to temperature changes.

For an absolute sensor under an open water surface, the reading includes atmospheric pressure plus water head. A barometric change of 1 hPa equals 100 Pa, approximately 0.40 in [10.2 mm] of water head. That can matter in a shallow tank.

Measure water level from a fixed datum and record the diaphragm elevation. For water near 68°F [20°C], a 1 in [25.4 mm] change in head produces approximately 249 Pa. Also record water temperature: density changes affect pressure at a fixed geometric depth.

Log these values together:

• Raw sensor output and converted pressure.
• Water level, diaphragm elevation and orientation.
• Water temperature and housing temperature.
• Atmospheric pressure when relevant to the sensor reference.
• Supply voltage and the state of nearby electrical equipment.

Compare the suspect sensor with a suitable independent pressure reference under stable conditions. Use manufacturer-approved calibration equipment rather than applying improvised pressure to a fragile diaphragm. Keep the original calibration unchanged while collecting diagnostic data.

Check deposits and mounting stress separately

Photograph the diaphragm and nearby metal before disturbing them. Record where the film accumulates, whether the pressure opening is restricted, and whether the housing has pits or powdery corrosion.

White deposits are not proof that aluminum or zinc has plated onto stainless steel. Tap water can leave mineral scale. Corrosion can produce insoluble compounds, and cathodic reactions can create local conditions that encourage precipitation. Laboratory analysis may be needed to distinguish these mechanisms.

A film may load a thin diaphragm or obstruct a pressure passage, depending on sensor construction. Its presence alone does not prove that it caused the drift. The same evidence-first approach appears in identifying residue inside an enclosure.

Do not scrape, polish or chemically strip the diaphragm. Ask the sensor manufacturer for an approved cleaning method and whether deposit samples should be retained. Cleaning before documenting the condition can remove useful evidence or damage the sensing element.

Next, compare the reading in the installed position with a manufacturer-approved, minimally stressed fixture at the same pressure and temperature. Check bore dimensions, alignment, insertion depth and seal compression against the sensor's installation requirements.

An O-ring can transmit mechanical load even without metal-to-metal contact. Water absorption, incompatible lubricant or dimensional changes can alter that load over time. Do not loosen a retaining feature while the assembly remains pressurized or submerged in service.

Verify coupling and run controlled comparisons

De-energize and isolate the assembly before resistance testing. Follow the site's electrical safety procedure and account for stored energy. Never use a resistance meter on an energized circuit.

Map possible electronic paths between:

• The stainless sensor body and aluminum or zinc housing.
• Sensor bodies sharing a common support or internal connection.
• The sensor case and cable shield.
• The housing and protective earth.
• Signal common, the power supply and connected instrumentation.

An insulating O-ring does not guarantee isolation. A retaining component, shield termination, shared instrument connection or protective conductor may complete the circuit elsewhere.

Measure continuity on accessible, isolated parts using the sensor manufacturer's permitted methods. A wet resistance reading may include electrochemical effects and does not reliably characterize metallic contact. Do not apply an insulation tester to sensor electronics unless the manufacturer expressly permits it.

Keep required protective bonding intact. The grounding and bonding guide explains why a protective fault-current path must not become a diagnostic variable. Never lift protective earth or remove a required bonding connection to see whether drift stops.

For controlled immersion comparisons, use a manufacturer-approved test fixture with a safe electrical arrangement. Change one factor at a time:

Record the complete wiring arrangement for every comparison. An insulating sleeve proves little if both metal parts remain connected through another path. If protective bonding prevents a valid isolation comparison, have the fixture redesigned; do not compromise the bond.

Use matched sensors where practical and repeat the comparison. A single successful soak does not establish a durable fix, particularly when the original drift developed slowly.

Match the fix to the confirmed cause

For deposits or housing corrosion: select wetted materials using the actual water chemistry, temperature and cleaning exposure. Consider a compatible polymer support or a suitably selected common-metal arrangement. Confirm diaphragm compatibility separately; changing the outer housing does not change the sensor's wetted materials.

For unintended electrical coupling: use an engineered isolation arrangement where permitted. Review sleeves, washers, retainers and electrical interfaces together. Preserve required protective bonding and verify that isolation survives assembly, immersion and maintenance.

For mounting stress: correct the bore, alignment, seal selection and retention method. Use the manufacturer's specified torque or retention procedure rather than additional compression to stop movement.

For reference-pressure problems: restore the documented vent arrangement or select a sensor designed for the intended submerged installation. Do not drill a reference opening or seal an existing vent without manufacturer approval.

For temperature effects: improve temperature measurement and installation stability before adding compensation. Software correction should address a characterized, repeatable effect, not conceal progressive deposits or damage.

If an offset remains after approved cleaning and stress-free reference testing, send the sensor for manufacturer evaluation. Re-zeroing alone does not establish that sensitivity, hysteresis or diaphragm integrity remains acceptable.

Stop the drift returning

Turn the successful comparison into an assembly specification. Record wetted materials, seal compounds, fit tolerances, electrical connections and the approved cleaning method. Include the reference-vent route where applicable.

Define acceptance limits before the next immersion test. Specify allowable offset, span change, hysteresis and recovery after removal. Set the exposure duration from the observed failure timing and intended service, not a convenient short soak.

Monitor water conductivity, pH, temperature and relevant contaminants. For tap-water trials, record the water source and replacement schedule. Evaporation and topping up can change dissolved-mineral concentration even when the visible level stays nearly constant.

Inspect both the housing and diaphragm at planned intervals. Do not assume a clean outer housing means the sensing surface is clean. The enclosure corrosion prevention article addresses the broader moisture and contamination controls around electrical equipment.

Retain an unaffected reference sensor and preserve the original wiring diagram. These make later changes easier to identify without repeating the entire investigation.

When to change the housing specification

Re-specify the assembly when drift returns after controlled corrections, isolation cannot survive maintenance, or the wetted materials are unsuitable. Re-specify it also when its depth, exposure duration or reference arrangement exceeds the documented design limits.

Separate two specifications: the enclosure protecting electrical equipment and the process housing carrying the pressure sensor. An electrical enclosure rating does not establish sensor accuracy, process-pressure capability or compatibility with a particular liquid.

NEMA 4X and IP66 do not establish suitability for continuous submersion. Custom Type 6P designs address prolonged submersion, but the specified depth, duration and complete assembly still require review. Our custom enclosure ratings are manufacturer ratings for the specified design; the enclosures are not UL listed.

For a custom enclosure specification, provide the liquid chemistry, temperature range, immersion depth and duration, wetted-material list, sensor drawing and required electrical connections. Include allowable drift and any listing requirements. Identify the pressure-sensor mounting interface explicitly rather than treating it as an ordinary cable cutout.

The useful outcome is a verified measurement assembly, not simply a more corrosion-resistant box.

Enclosures for this job

Catalog sizes that suit the job this article answers, picked from its subject; open one for its full specifications.

Questions