What Is Condition Monitoring?
Condition monitoring is the practice of measuring and tracking the physical parameters of operating equipment — vibration, temperature, lubricant condition, ultrasonic emissions, electrical characteristics — to detect changes that indicate developing faults. It is the data collection and diagnostic engine that makes predictive maintenance possible. Without condition monitoring, your maintenance program is operating blind, relying on calendar schedules and operator intuition to decide when equipment needs attention.
A condition monitoring program is more than a collection of instruments. It is a systematic process that includes defining what to monitor, selecting the right technology for each failure mode, establishing measurement points and data collection procedures, setting alarm thresholds, analyzing trends over time, diagnosing fault conditions, and communicating findings to the people who schedule and execute maintenance work. Each of these steps must be executed consistently for the program to deliver value. A vibration sensor on a bearing housing is worth nothing if the data isn’t collected on schedule, analyzed by someone who understands the machine, and acted on by a maintenance team that trusts the findings.
Modern condition monitoring services span a spectrum from periodic route-based data collection using portable instruments to permanently installed continuous monitoring systems that stream data to cloud-based analysis platforms. The right approach depends on equipment criticality, failure consequences, operating speed, accessibility, and budget. Most industrial facilities benefit from a blended strategy — continuous monitoring on the highest-criticality assets that justify the investment, and route-based monitoring on the broader equipment population where periodic data collection provides adequate warning.
How a Multi-Technology Monitoring Program Works as a System
Individual monitoring technologies are powerful, but they become significantly more effective when deployed as an integrated system. Each technology has strengths and limitations defined by the physics of what it measures. Vibration analysis is the workhorse for rotating equipment — it detects imbalance, misalignment, bearing defects, gear mesh anomalies, resonance, and looseness with high specificity. But vibration analysis is less effective on equipment operating below approximately 100 RPM, has limited sensitivity to lubricant degradation, and cannot detect electrical insulation breakdown.
Oil analysis fills several of those gaps. Wear metal trending through spectrometric analysis detects active component wear often before vibration signatures change. Particle counting and analytical ferrography characterize the type, size, and severity of wear debris. Fluid chemistry tests — acid number, base number, oxidation, water content — reveal lubricant degradation that leads to accelerated wear if left uncorrected. For slow-speed equipment like paper machine rolls, kiln support rollers, and large gear drives, oil analysis is frequently the primary or only effective monitoring technology.
Infrared thermography detects thermal anomalies that are invisible to vibration or oil analysis. Loose or corroded electrical connections, overloaded circuit breakers, failing capacitor banks, refractory deterioration in kilns and furnaces, blocked heat exchanger tubes, and failed steam traps all produce detectable thermal signatures. Thermographic surveys of electrical distribution systems are one of the highest-return condition monitoring activities available, with insurance industry data indicating that routine IR surveys can prevent 30-50% of electrical fire incidents in industrial facilities.
Insurance industry data indicates that routine infrared surveys can prevent 30-50% of electrical fire incidents in industrial facilities.
Ultrasonic monitoring — both airborne and structure-borne — detects high-frequency phenomena that other technologies miss. Compressed gas and vacuum leaks produce ultrasonic emissions that can be located and quantified with handheld instruments. Steam trap surveys using ultrasound are faster and more accurate than temperature-based methods. Early-stage bearing lubrication issues — where metal-to-metal contact produces high-frequency stress waves before measurable vibration develops — are detectable with ultrasonic monitoring months before conventional vibration analysis would flag a problem. Partial discharge in medium- and high-voltage electrical insulation systems generates ultrasonic signals that provide early warning of insulation breakdown.
When these technologies operate as a coordinated system, with findings from each technology informing the analysis of the others, the result is a monitoring capability that is both broader (covering more failure modes) and deeper (providing higher diagnostic confidence) than any single technology can achieve alone.
What Are the Signs Your Facility Needs Condition Monitoring Services?
Condition monitoring delivers value across a wide range of facility types and maturity levels. Whether you’re starting from scratch or looking to strengthen an existing program, the following indicators suggest that professional condition monitoring services would improve your equipment reliability and reduce your maintenance costs.
- Your facility has no systematic method for assessing equipment health between scheduled maintenance events — you are essentially operating until something breaks or an operator notices a problem
- Operators report abnormal noise, vibration, or heat from equipment, but there is no baseline data to determine whether conditions are actually deteriorating or within normal range
- You have attempted to build an in-house monitoring program but lack the trained analysts to interpret data and generate actionable diagnoses
- Equipment failures are occurring without warning, leading to unplanned downtime, emergency repairs, and cascading production impacts
- Your preventive maintenance program includes intrusive inspections (bearing replacements, coupling checks, alignment verifications) performed on arbitrary time intervals, and you want to shift to condition-based triggers
- Critical equipment is inaccessible during normal operation due to safety zones, enclosures, or process temperatures, making visual or manual inspection impractical
- Insurance carriers have recommended or required equipment monitoring as a condition of coverage, particularly for high-value rotating equipment or electrical systems
- Your facility operates on a continuous or semi-continuous schedule where unplanned shutdowns carry production losses of $10,000 per hour or more
- You have recently acquired or inherited a facility and have limited knowledge of the installed equipment’s current condition or maintenance history
- Your current monitoring program exists but has gaps — perhaps only vibration is covered, and you suspect failure modes related to lubrication, electrical systems, or thermal conditions are being missed
Our Condition Monitoring Approach
We design and execute condition monitoring programs that produce clear, actionable information — not just data. The difference is significant. A vibration spectrum is data. A diagnosis that says “the outboard bearing on pump P-301A has an outer race defect in stage 2 severity, estimated remaining life is 60-90 days at current load, recommend bearing replacement during the next planned outage on April 15th” is information that your planning team can act on immediately.
Technology Selection Matrix
We select monitoring technologies based on a structured evaluation of the equipment type, dominant failure modes, operating conditions, and consequence of failure. This technology selection matrix ensures that every monitoring point has a defined purpose tied to a specific failure mode, and that no significant failure modes are left unmonitored due to reliance on a single technology.
For a typical centrifugal pump in critical service, the technology selection might include tri-axial vibration measurements at each bearing housing (targeting imbalance, misalignment, bearing defects, and looseness), quarterly oil sampling with wear metal, particle count, and fluid chemistry analysis (targeting lubrication degradation and incipient bearing wear), and annual thermographic survey of the motor and junction box connections (targeting electrical connection integrity). For a large forced-draft fan, the selection might add continuous vibration monitoring with automated alerts, displacement probes at the bearing pedestals, and ultrasonic monitoring of bearing lubrication conditions.
The matrix is not static. As the program matures and failure history accumulates, we adjust technology assignments based on what the data reveals. If a particular equipment class is not generating findings through vibration monitoring but is producing meaningful wear metal trends through oil analysis, we may reduce the vibration collection frequency and increase the oil sampling frequency — optimizing both cost and detection effectiveness.
Continuous vs. Route-Based Monitoring
The decision between continuous online monitoring and periodic route-based data collection is driven by a combination of criticality, P-F interval, accessibility, and economics. Continuous monitoring is justified when the consequence of failure is extreme (safety-critical or high-production-impact equipment), the P-F interval is short (meaning rapid deterioration is possible once a fault initiates), or the equipment is inaccessible for routine data collection due to safety barriers, confined spaces, or remote location.
Route-based monitoring remains the most cost-effective approach for the majority of industrial equipment. A trained analyst collecting vibration data monthly on a centrifugal pump with a bearing P-F interval of three to nine months has ample detection margin. The economics are straightforward: a permanently installed wireless vibration sensor on a single bearing point may cost $500-$2,000 for hardware plus ongoing connectivity and platform fees, while a monthly route-based measurement at the same point costs a fraction of that annually. When you multiply across hundreds of monitoring points, the cost advantage of route-based monitoring for non-critical equipment is substantial.
Our programs typically deploy continuous monitoring on 5-15% of the asset base — the equipment where failure consequences are highest — while covering the remaining 85-95% with disciplined route-based collection.
Alarm Thresholds and Trending
Setting appropriate alarm thresholds is one of the most consequential — and most frequently mishandled — elements of a condition monitoring program. Thresholds that are too loose generate false confidence; equipment can progress from healthy to failed without ever triggering an alert. Thresholds that are too tight produce excessive false alarms, eroding the maintenance team’s trust in the program and driving “alarm fatigue” behavior where warnings are routinely dismissed.
We set alarm thresholds using a layered approach. Absolute thresholds based on industry standards — ISO 10816 for vibration severity on broad equipment classes, machine-specific OEM specifications where available — provide a baseline. Equipment-specific statistical thresholds, calculated from the asset’s own historical data, provide a more sensitive and accurate alert level once sufficient operating history has been collected. Trending thresholds — alerts triggered not by the current value but by the rate of change — catch developing faults that are still below absolute alarm levels but are accelerating in a way that indicates a worsening condition.
CMMS Integration
Condition monitoring data that lives in isolation — in a separate database or on a standalone workstation — is condition monitoring data that gets ignored. We structure our programs to integrate findings directly into our clients’ computerized maintenance management systems. When a monitoring analyst identifies a developing fault, the recommended corrective action is entered as a work request in the CMMS, complete with diagnosis, severity assessment, recommended intervention window, and required parts and resources. This integration ensures that condition-based findings enter the same planning and scheduling workflow as all other maintenance work, eliminating the gap between “we found a problem” and “we fixed the problem” that undermines too many monitoring programs.
What Equipment Is Typically Covered?
Pumps and Compressors
Centrifugal, reciprocating, and screw-type pumps and compressors across all size ranges and services. Monitoring typically includes vibration analysis for mechanical faults, oil analysis for lubrication health and wear metal trending, and ultrasonic monitoring for cavitation detection in pumps and valve leakage in reciprocating compressors. Large centrifugal compressors in critical service often warrant continuous proximity probe monitoring in accordance with API 670 requirements.
Fans, Blowers, and Air Handling Equipment
Induced-draft and forced-draft fans in combustion processes, process gas blowers, HVAC air handling units in pharmaceutical and semiconductor manufacturing, and cooling tower fans. These assets are particularly susceptible to imbalance from buildup accumulation, corrosion, and erosion — conditions that vibration monitoring detects reliably. Large ID fans in power generation and cement manufacturing often require both casing-mounted accelerometers and shaft proximity probes to fully characterize rotor behavior.
Electric Motors
From fractional horsepower to tens of thousands of horsepower, electric motors drive the majority of industrial rotating equipment. Monitoring coverage includes vibration analysis for bearing condition, balance, and alignment, infrared thermography for connection integrity and winding temperature, motor current analysis for rotor bar integrity and supply-side anomalies, and insulation resistance testing for stator winding health. For large motors above 500 HP in critical service, motor circuit analysis provides offline diagnostic capability that can detect developing insulation faults, connection resistance changes, and rotor asymmetry well in advance of failure.
Gearboxes and Drive Systems
Parallel shaft, planetary, and worm gear reducers in all industrial applications. Gearboxes are among the highest-value assets in most facilities, and their internal components — gears, bearings, shafts, seals — are inaccessible for visual inspection during operation. Vibration analysis with high-resolution spectral data is the primary technology for gear condition assessment, complemented by oil analysis for wear metal trending and lubricant condition. Acoustic emission monitoring is increasingly used on large, slow-speed gearboxes where conventional vibration analysis lacks sensitivity.
Electrical Distribution Systems
Switchgear, transformers, bus ducts, cable terminations, and motor control centers. Infrared thermographic surveys — conducted under load conditions representative of normal operation — are the primary monitoring technology, supplemented by ultrasonic partial discharge surveys on medium-voltage systems. Transformer monitoring may additionally include dissolved gas analysis of insulating oil, which can detect thermal faults, arcing, partial discharge, and cellulose degradation within the transformer.
Steam Systems and Heat Exchangers
Steam traps, pressure-reducing valves, condensate return systems, and shell-and-tube or plate heat exchangers. Ultrasonic steam trap surveys can assess hundreds of traps per day with high diagnostic accuracy — the U.S. Department of Energy estimates that a typical steam system has a 15-30% trap failure rate, and each failed trap can waste thousands of dollars in steam energy per year. Heat exchanger performance monitoring through approach temperature and pressure drop trending detects fouling and flow maldistribution that degrades process efficiency.
The U.S. Department of Energy estimates that a typical steam system has a 15-30% trap failure rate, and each failed trap can waste thousands of dollars in steam energy per year.
Source: U.S. Department of Energy
What Results Do Companies Typically See?
A well-executed condition monitoring program produces results that are measurable at the equipment level, the department level, and the facility level. The following outcome ranges reflect what we consistently observe across different industries and starting conditions.
- Fault detection rate of 85-95% for monitored equipment — the percentage of failure events that are detected by the monitoring program before functional failure occurs, allowing planned intervention
- Unplanned downtime reduction of 35-55% on monitored equipment as developing faults are identified and corrected during planned maintenance windows rather than through emergency response
- Maintenance cost reduction of 20-30% from avoided emergency repairs, reduced overtime, lower expedited parts procurement costs, and elimination of unnecessary preventive replacements
- Energy loss recovery of 8-15% in compressed air and steam systems through systematic leak detection and steam trap failure identification — losses that are invisible without ultrasonic monitoring
- Mean time between failures (MTBF) improvement of 25-40% as condition data informs precision maintenance practices and operating adjustments that extend component life
- Spare parts inventory optimization of 10-20% as failure prediction replaces safety stock for emergency scenarios, and procurement lead times can be aligned with actual need dates rather than worst-case assumptions
- Insurance premium reductions in some cases, as carriers recognize the risk reduction associated with systematic equipment monitoring programs — particularly for electrical systems and high-value rotating equipment
The compounding effect of condition monitoring is worth emphasizing. In the first year, the primary benefit is early fault detection — catching problems before they cause unplanned downtime. By the second and third year, the accumulated data enables trending-based predictions with longer lead times, better procurement planning, and more efficient outage scheduling. By year three to five, the historical database supports statistical analysis that can identify fleet-wide failure patterns, vendor quality issues, and design deficiencies that a shorter dataset would not reveal.
Building a condition monitoring program is an investment in visibility. You cannot manage what you cannot measure. Our team brings the instrumentation, the analytical expertise, and the systematic discipline to give you clear, continuous visibility into the health of the equipment your operation depends on — and to translate that visibility into maintenance decisions that protect your production, your people, and your profitability.