Nuisance oil alarms from DOCA sensors

A DOCA sensor can raise an oil-contamination alarm even when the compressed air contains no actual oil. Troubleshooting this behaviour starts with treating the alarm as evidence rather than a verdict: condensate, residue, baseline drift or a processing fault can produce a nuisance alarm in liquid, aerosol or vapour detection modes.

Work in a controlled order: confirm the alarm mode, reproduce the event with approved reference air, compare raw and processed readings where available, and preserve the configuration before making changes. This protects sample integrity and gives the DOCA project useful diagnostic evidence.

Decode the alarm before touching the sensor

Start by classifying the event as liquid, aerosol or vapour detection and checking whether the alarm is instantaneous, latched or repeated. A transient puff can be more significant than a steady trace, while a latched alarm may preserve a peak long after the air has returned to normal.

Review the event log, trend history and commissioning records for matching spikes. Pair each alarm with pressure, flow, temperature, humidity and maintenance activity. Never reset repeatedly before capturing the state, and do not weaken a threshold simply because tests appear clean.

Rule out artefacts in the sample line

Sample lines can introduce the very material the optical sensor is meant to detect. Condensed oil, moisture, corrosion products, fingerprints or cleaning-agent residue may scatter light or contaminate the measurement surface. Incorrectly fitted or ageing filters, sample drains and relief valves also change what reaches the sensor.

Follow the manufacturer’s approved purge, blanking and zeroing procedure with the process isolated or placed in a safe test state. Have a look for a recently opened sample path, a swapped filter or technician residue. Repeat the test with a clean, uncontaminated sample and compare the raw signal.

Run a controlled false-alarm diagnostic checklist

Use this checklist while the line is isolated or in the safe operating state defined by the site risk assessment. It prevents a quick reset from erasing useful evidence.

Bench-side checks

  • Capture the timestamp, alarm mode, raw signal, processed value and line conditions before resetting.
  • Run the approved zero or reference-air check using a known oil-free supply.
  • Inspect the optical window and sample connections for condensation, dust, damage or residue.
  • Confirm filter orientation, replacement date and the condition of drains or relief devices.
  • Compare the sensitivity and alarm threshold with the approved commissioning baseline.

At the plant, repeat the check across normal operating demand and compare it with a verified oil-free branch or supply. Coordinate the test window with production, maintenance and quality personnel.

Site-side checks

  • Test the sensor on an oil-free supply to separate a sensor fault from a process event.
  • Repeat measurements at minimum, nominal and maximum flow or pressure.
  • Check the compressor, seals, filters and recent maintenance for lubricant carry-over.
  • Correlate alarms with nearby equipment that may release oil mist or hydrocarbon vapour.
  • Record every intervention and repeat the alarm test before returning the line to service.

Change one variable at a time and record the result. If the raw signal shifts, focus on sample condition and optics; if it stays flat, examine alarm logic, calibration and configuration.

Separate optical faults from process contamination

An optical channel can be stable while the sampled air contains genuine oil, and non-oil dust can imitate the target aerosol. First determine whether the raw response moved. Inspect condensation, stray light, vibration, damaged connectors and electrical interference if the signal changes without a corresponding process event.

Other hydrocarbons and solvent residues can also create a positive vapour response. Cross-sensitivity, incomplete baseline subtraction and temperature-compensation errors should be checked against the approved method, not guessed from a single reading.

The separation of optics, mechanics and software used in machine vision integration offers a useful analogy: validate each layer before blaming the complete system.

Check calibration, thresholds and data integrity

Once physical causes are excluded, review the zero baseline, sensitivity, alarm threshold, hysteresis and any unit conversion applied by the controller. Firmware changes or a replaced configuration file can create a spurious trip without changing the air quality.

Use the manufacturer’s specified calibration gas or traceable reference procedure. Do not improvise with an oil spray, workshop aerosol or compressed-air duster; these can contaminate the line and invalidate the result. Record all authorised parameter changes and repeat the alarm test before sign-off.

Adapt troubleshooting to Australian industrial conditions

Australian sites range from pharmaceutical and medical-gas facilities around Melbourne and Adelaide to electronics plants in Sydney and biotech operations in Queensland. Local teams often say they will “have a look” at a flagged unit, then “sort it out”; use that plain language in maintenance notes without losing technical detail.

Heat and humidity can encourage condensate, inland dust can foul an optical window, and coastal salt can corrode connectors. The DOCA technology also comes from a European R&D programme, so Australian buyers should confirm local technical support, spare parts, firmware access and how patent protection or licensing would work before deployment.

For traceable calibration, use an Australian laboratory accredited by NATA, and apply isolation and lockout/tagout controls consistent with Safe Work Australia principles. These local checks do not replace the manufacturer’s DOCA-specific procedure.

Escalate the issue and restore a reliable baseline

Restore confidence through a staged return to service: clean the sample path, establish a stable zero, repeat the reference test under normal and peak demand, and obtain quality sign-off. Never bypass or broaden an alarm without authorised change control, especially in pharmaceutical, hospital and clean-room settings.

If the nuisance event remains, contact the DOCA project team through the website with the alarm log, raw and processed traces, configuration, sample-line diagram and maintenance history. That evidence enables root-cause analysis, supports design and patent feedback, and helps the affected Australian facility restore a defensible oil-free baseline without sacrificing contamination protection.