Best practices for maintaining and calibrating online process analyzers
Online process analyzers provide continuous insight into treatment performance. Measurements from pH, dissolved oxygen, ammonia, nitrate, chlorine, turbidity, conductivity, and oxidation-reduction potential instruments can guide chemical dosing, aeration, process control, compliance reporting, and early fault detection.
Their value depends on more than selecting a reliable instrument. Fouling, air bubbles, changing flow conditions, expired standards, poor sample conditioning, and unnoticed sensor drift can turn a live trend into misleading information. A disciplined maintenance and calibration program keeps analyzer readings connected to actual process conditions.
Water and wastewater professionals across the Los Angeles area share practical knowledge through LABS of CWEA, including technical education, facility tours, and workshops. The same habits emphasized in those settings—careful observation, consistent procedures, and clear records—support dependable online monitoring.
Start with the process and installation
An analyzer should be installed where its sample represents the process being managed. Avoid stagnant zones, locations immediately downstream of chemical injection points, and areas with excessive turbulence or sediment accumulation unless the instrument is specifically designed for those conditions. For sidestream systems, confirm that the flow rate, pressure, temperature, and return point match the analyzer manufacturer’s requirements.
Sample lines need special attention. Long or narrow tubing can create lag, adsorption, air entrapment, and solids deposition. Use suitable materials, minimize unnecessary bends, and provide isolation valves and drains that allow safe service. Flow indicators, strainers, sample conditioning systems, and automatic cleaning devices should be accessible rather than hidden behind unrelated equipment.
Record the installation details in the asset management system. Include the analyzer model, sensor type, process location, sample flow range, communication method, alarm limits, and maintenance interval. This information helps operators distinguish a true process change from an installation problem.
Establish a risk-based maintenance routine
Maintenance frequency should reflect the measurement, process, and consequences of an incorrect reading. A chlorine analyzer in a disinfection control loop may need more frequent attention than a conductivity instrument used for general trending. A wastewater stream with grease, grit, biological growth, or high suspended solids will usually require more cleaning than a relatively clear utility-water sample.
Routine inspections should look for fouled membranes, cracked housings, damaged cables, blocked sample lines, depleted reagents, leaks, loose fittings, and unstable flow. Review the displayed value alongside the signal quality, diagnostic codes, temperature compensation, and recent trend. Sudden flatlining, excessive noise, or a reading that changes only after manual intervention can indicate an instrument issue.
Cleaning methods must suit the sensor. Aggressive brushes, unsuitable solvents, or abrasive materials can damage sensitive surfaces and shorten sensor life. Follow the manufacturer’s procedure, rinse thoroughly, and allow the instrument to stabilize before judging its response. Where automatic cleaning is installed, verify that spray pressure, air supply, wipers, and cleaning chemicals are actually functioning.
Calibrate with traceable standards
Calibration adjusts an analyzer to a known reference, while verification checks whether the instrument still agrees with that reference. Both activities matter. A calibration performed with contaminated standards or an incorrectly prepared solution can make the instrument appear accurate while shifting its measurement away from reality.
Use standards within their expiration date and appropriate concentration range. Label opened containers with the opening date, protect them from contamination, and never return used solution to the original bottle. For pH, use fresh buffer solutions and allow the sensor to reach a stable temperature. For chlorine, ammonia, nitrate, and similar measurements, follow the specified reagent preparation, reaction time, and sample temperature requirements.
The number of calibration points should match the analyzer and its operating range. A two-point pH calibration is common, while other instruments may require a zero and span procedure or a multi-point curve. Record the standard values, lot numbers, temperature, as-found reading, as-left reading, operator, and any corrective work. This record establishes whether drift is gradual, sudden, or associated with a particular process condition.
| Analyzer type | Common reliability risks | Useful routine checks | Calibration or verification focus |
|---|---|---|---|
| pH and ORP | Coating, aging reference junction, temperature effects | Clean sensor, inspect junction, confirm stable flow | Fresh buffers or reference solution; allow stabilization |
| Dissolved oxygen | Membrane damage, bubbles, biological growth | Inspect membrane, cable, flow, and air exposure | Air-saturated water or approved zero solution |
| Chlorine | Reagent depletion, flow variation, pH interference | Check reagents, sample line, temperature, and flow | Zero and span checks with suitable standards |
| Ammonia and nitrate | Fouling, reagent errors, matrix effects | Inspect tubing, reagents, filters, and sample conditioning | Multi-point check or laboratory comparison |
| Turbidity and suspended solids | Optical window fouling, bubbles, color interference | Clean optics, remove bubbles, inspect alignment | Formazin or approved traceable standards |
| Conductivity | Coating, incorrect cell constant, temperature errors | Clean cell, inspect compensation, confirm immersion | Certified conductivity standard at controlled temperature |
Compare readings before changing the process
An online value should be evaluated against an independent reference when it drives an important decision. Collect a properly preserved grab sample at a representative location and compare it with a laboratory result or a calibrated portable instrument. The comparison must account for sample transport time, holding limits, temperature, analytical method, and the physical distance between sampling points.
A difference does not automatically prove analyzer failure. Online and laboratory samples may represent different moments, especially in a fast-moving process. Poor mixing, delayed sample transport, or a sidestream lag can explain part of the variation. Repeated bias in the same direction, however, warrants investigation of calibration, sample conditioning, sensor age, and laboratory method performance.
Set practical acceptance criteria before a problem occurs. These may be expressed as an absolute difference, a percentage difference, a control limit, or a process-specific operating band. When a result falls outside the criterion, document the investigation and identify whether the corrective action is cleaning, recalibration, sensor replacement, sample-system repair, or process review.
Protect data quality and operator decisions
An analyzer can be technically functional while producing data that should not be used. Mark readings as questionable during calibration, cleaning, reagent replacement, sample-line flushing, power interruptions, or communication failures. Supervisory control and data acquisition systems should preserve these flags rather than presenting all values as equally reliable.
Alarm settings need regular review. A narrow alarm band may generate nuisance alerts when normal process variability is high, while a broad band can delay detection of a real upset. Use time delays, rate-of-change alarms, and redundant measurements where appropriate. Operators should know what action follows each alarm and when a manual sample is required.
Keep maintenance and calibration records linked to the analyzer tag. Trend the as-found error, downtime, failed checks, replacement parts, and recurring alarms. These records can reveal that a sensor is being cleaned too often, that a sample pump is wearing out, or that a particular installation location consistently causes fouling. Data quality improves when reliability information is reviewed alongside process performance.
Build a practical team standard
Consistent procedures matter when several shifts, contractors, and departments share responsibility for online instrumentation. Create short, site-specific work instructions with photographs or diagrams for isolation, cleaning, calibration, sample collection, waste handling, and return to service. Include the required personal protective equipment and lockout or chemical safety steps.
Training should cover the measurement principle as well as the button sequence on the instrument. An operator who understands how temperature, bubbles, ionic strength, reagents, or biological fouling affect a reading is better equipped to recognize a questionable result. Professional events and CWEA news and updates can also help teams stay aware of technical workshops, facility practices, and emerging approaches to water quality monitoring.
Use the following actions to turn general guidance into a manageable program:
- Assign an owner and backup for every critical analyzer.
- Set inspection, cleaning, calibration, and verification intervals in the maintenance system.
- Keep certified standards, replacement parts, and documented procedures available at the point of work.
- Compare critical online measurements with independent results at a defined frequency.
- Review analyzer performance and recurring failures during routine operations meetings.
Make reliability visible in daily operations
The strongest programs treat analyzer maintenance as part of process control rather than as an isolated instrumentation task. Operators, laboratory staff, electricians, mechanics, engineers, and supervisors should share a common understanding of what the measurement means and what to do when it becomes questionable.
Begin with the analyzers that influence permit compliance, chemical feed, aeration, worker safety, or public health. Audit their installation, maintenance history, calibration records, alarm response, and independent verification results. Then apply the lessons to lower-risk instruments, using failure data to refine inspection intervals and spare-parts planning.
Review your analyzer register, select one critical measurement, and document its next inspection and verification before the next shift change. A clear procedure, a traceable standard, and a reliable comparison sample can turn continuous measurement into information your team can confidently use.