0
October 19, 2026

Beyond the Noise: Building an Effective Airborne & Structure-Borne Ultrasound Condition Monitoring Program

Predictive maintenance (PdM) programs rarely struggle because an organization lacks technology. More often, they struggle because the technology never becomes fully integrated into the way the organization manages its assets, maintenance activities, and operational risk. Airborne & Structure-borne ultrasound instruments and sensors can detect compressed air leaks, identify changes in bearing friction, assist with condition-based lubrication, detect electrical discharges, evaluate the condition of valves and steam traps, and provide insight into numerous other asset conditions. However, purchasing an instrument, training a technician, or even finding defects does not necessarily mean an organization has developed an effective ultrasound predictive maintenance program.

The real value of condition monitoring is created when the information from PdM technologies like ultrasound becomes part of a repeatable process for identifying potential failures early enough to make better maintenance and business decisions. That distinction is at the heart of building a mature ultrasound program. Organizations should eventually move beyond asking, “What can we find with ultrasound?” and begin asking a much more important question: How can the information generated by ultrasound help us improve asset reliability, maintenance execution, operational performance, and ultimately the performance of the business? 

That is the journey beyond the noise. 

Start With Reliability, Not Technology 

One of the easiest mistakes to make when implementing any predictive maintenance technology is allowing the technology itself to become the program. An organization purchases an ultrasound instrument, someone attends a training class, compressed air leaks are identified, bearings are inspected, and perhaps an electrical anomaly is discovered. These are all positive outcomes, but they represent activities rather than a comprehensive reliability strategy. 

A mature condition monitoring program begins by understanding the organization’s reliability objectives. What assets create the greatest operational risk? Which failures have the greatest consequences? Where is unplanned downtime affecting production? Which failure modes can reasonably be detected using ultrasound? Most importantly, where can earlier detection provide maintenance with enough time to plan and schedule corrective work rather than responding to an emergency? 

Answering these questions changes the conversation. Instead of simply deciding where ultrasound can be used, the organization begins determining where ultrasound should be used. This is where knowing what assets are critical becomes important because not every asset deserves the same level of monitoring. Resources should be concentrated where failures have meaningful consequences to safety, production, quality, environmental performance, cost, or other organizational objectives. Ultrasound then becomes part of the larger asset reliability strategy rather than an isolated PdM activity. 

Establishing the Foundation 

Early-stage ultrasound programs should start relatively small and simple. The initial objective should not be to monitor every possible asset or application simply because the technology is capable of doing so. Trying to accomplish too much too quickly can result in enormous inspection routes, excessive amounts of data, and more findings than the maintenance organization can realistically address. A better approach is to begin with a manageable population of critical assets and applications where ultrasound can provide actionable information.

Depending on the facility, these applications might include critical motors, pumps, bearings, electrical equipment, compressed air systems, steam trap testing, valve inspections, or other assets where ultrasound provides an appropriate inspection method. The organization should also establish basic standards for collecting data. For mechanical condition monitoring, this means establishing repeatable and consistent measurement locations and consistent inspection techniques. Baseline measurements should be established when possible so future readings can be evaluated against historical conditions rather than viewed as isolated measurements. Baseline ultrasound data should include both static (dB and Crest Factor) and dynamic (sound file recordings) data for trending and fault analysis.   

With any PdM technology, consistency is especially important because variations in measurement technique can create variations in data that have nothing to do with the actual condition of the asset. A reading collected from one location today and a different location several months later will make trending difficult or even misleading. The same principle applies to sensor placement, operating conditions, instrument settings, inspection intervals, and data collection procedures. The goal is to create repeatable information that can be compared over time.  With the capabilities of today’s ultrasound instruments, much of this can be defined in the ultrasound instrument’s software and associated database software to help guide the technician in the data collection process.   

Training is equally important during this stage. Technicians need to understand more than how to operate the instrument. They should understand the assets being inspected, common failure modes, appropriate applications for ultrasound, the limitations of the technology, and the science & theory behind it. Technology provides the measurement, but for route-based ultrasound inspections, people must provide context and interpretation. A successful program; therefore, develops both technical capability and asset knowledge. 

When first starting out, many developing ultrasound programs initially operate in a “find the bad actor” mode. An inspector walks through the facility looking and listening for something that appears different. This can produce valuable discoveries, particularly during troubleshooting activities.  This “comparison method approach” also helps to identify potential problems early, thus creating some quick wins that help for buy-in and quick ROI.  As the program evolves and matures, the predictive maintenance program depends on a more historical or trending approach in order to understand how an asset’s condition changes over time. 

Consider a bearing inspection. A single elevated ultrasound dB reading may justify additional investigation, but the historical trend of that bearing can provide significantly more context. Has the ultrasound level gradually increased over the previous six months? Did it increase rapidly during the most recent inspection interval? Did the condition improve or deteriorate further following a lubrication event? Are there corresponding changes in vibration, temperature, operating load, speed, or process conditions? A trend can tell a story that an isolated measurement cannot. 

This is where ultrasound begins transitioning from a troubleshooting tool into a true condition monitoring technology. Routes become defined, measurement locations are standardized, inspection intervals become risk-based, baseline data is established, alarm criteria become more meaningful, and findings are documented in a consistent manner. The organization is no longer simply collecting ultrasound readings; it is developing an asset condition history that becomes increasingly valuable as the program matures.  More importantly, PdM data is used to make data-driven decisions on asset condition and repairs, versus guessing or performing maintenance tasks simply on time-based intervals alone.   

Connecting Condition Monitoring to Work Management 

One of the most important—yet sometimes overlooked—steps in predictive maintenance maturity is connecting condition monitoring findings to the maintenance work management process. Finding a potential failure has limited value if nothing meaningful happens afterward. The ultimate objective is not simply to identify problems; it is to create enough information and time for the organization to act before those problems become functional failures.

Suppose ultrasound identifies a developing bearing condition several weeks before the bearing reaches functional failure. That discovery creates an opportunity, but the value of the opportunity depends entirely on what happens next. The finding may need to be validated, documented, prioritized, converted into a corrective work order, and incorporated into the maintenance backlog. Parts may need to be identified and ordered, labor scheduled, and the repair coordinated with production.

If these processes are disconnected, the predictive maintenance program may successfully detect developing failures while the organization continues experiencing emergency maintenance. 

This illustrates one of the most important concepts in predictive maintenance: the real product of early detection is time. The earlier the fault is detected, increases the amount of time available between identifying a potential failure and reaching functional failure. That additional time gives maintenance and operations more options. Instead of responding to an unexpected failure in the middle of the night, maintenance may be able to schedule the repair during a planned outage. Instead of expediting parts, the organization may use normal procurement channels. Instead of paying overtime or using contractors, the work may be completed during normal working hours. Instead of unexpectedly losing production, operations may determine the least disruptive time to release the equipment for maintenance. 

At this point, ultrasound has moved beyond inspection and into execution. The technology is no longer valuable merely because it detects a condition; it is valuable because the information changes how work is planned, scheduled, and performed. 

Looking Beyond the Traditional P-F Curve 

The traditional P-F curve provides a useful framework for understanding the role that predictive maintenance plays in an asset’s operation. Point P represents the point at which a potential failure becomes detectable, while Point F represents functional failure. The interval between those two points represents the opportunity available to detect, diagnose, plan, schedule, and correct the condition before the asset can no longer perform its required function.

However, organizations seeking greater reliability should also consider what happens before Point P. Many failures have origins in design, installation, commissioning, operation, lubrication practices, contamination control, alignment, electrical connections, or other upstream conditions. This creates a broader S-D-I-P-F perspective: Specify, Design, Installation, Potential Failure, and Functional Failure. 

Specifying the right assets that deliver the desired throughput and production levels needed for the organization’s goals should be discussed before being procured.  Design influences and designing for reliability is appropriate for the application and operating environment, and designing for reliability and maintainability.  Installation influences whether the asset begins its operating life under favorable conditions, helping to extend the amount of time between Point I and Point P. Point P represents the stage at which degradation has progressed enough to become detectable with a PdM technology, while Point F represents the point at which the asset can no longer perform its required function that it was designed and installed for. Looking at the asset lifecycle this way reinforces an important reliability principle: the best failure is not simply the failure that is detected early; the best failure is the failure prevented from developing in the first place. 

Condition monitoring remains essential, but a mature reliability program within an organization uses condition monitoring findings to improve the upstream processes that create those findings. If ultrasound repeatedly identifies lubrication-related bearing problems, for example, the long-term solution should not simply be to become better at finding poorly lubricated bearings. The organization should investigate why those lubrication conditions continue to occur. Lubrication intervals may be incorrect, technicians may be applying too much or too little grease, the wrong lubricant may be used, contamination controls may be inadequate, or storage, delivery, and filtration practices (or lack thereof) may be contributing to premature degradation. 

This represents an important transition in program maturity. Condition monitoring begins with defect detection, but mature reliability programs use what they learn from those defects to pursue defect elimination through Root Cause Analysis.  All in an effort to prevent those failures from happening again.   

Measuring Effectiveness Instead of Activity 

Another indication of program maturity is a change in how success is measured. Developing programs often focus heavily on activity metrics such as the number of assets inspected, routes completed, compressed air leaks identified, or reports generated. These measurements can help determine whether the program is being executed, but they do not necessarily demonstrate whether reliability is improving. 

A technician can inspect hundreds of assets and generate dozens of reports without preventing a single functional failure. Mature programs begin shifting from measurements of activity toward measurements of effectiveness and outcomes. Organizations should understand how many actionable conditions are being identified, how much corrective work originates from predictive maintenance findings, what percentage of those findings are corrected before functional failure, and whether predictive maintenance is contributing to a reduction in emergency and reactive work. 

The organization should also examine broader indicators such as planned work, percent of planned work from PdM finds, schedule compliance, Mean Time Between Failures, availability, asset utilization, repeat failures, and other measures that connect maintenance execution with operational performance. The objective should not be to demonstrate that the ultrasound program is busy. The objective should be to demonstrate that the effectiveness of the ultrasound program and that the organization is becoming more reliable.  

The Power of Marginal Gains 

Reliability improvement rarely occurs because of one dramatic initiative. More often, it is the result of dozens of smaller improvements that are executed consistently over time. A bearing failure prevented here, a major compressed air leak repaired there, a lubrication interval optimized, an electrical anomaly corrected during a scheduled outage, failed steam traps identified, a recurring defect eliminated, or a better installation practice implemented may each appear relatively small when viewed independently. Collectively, however, these improvements can significantly change the performance of an operation. 

Consider a facility capable of producing $100 million of annual output at its current operating rate. If improvements in reliability enable the organization to increase effective asset utilization by only one percent, the resulting production opportunity could represent approximately $1 million in additional annual output, assuming sufficient demand and no other production constraints prevent that capacity from being utilized. A two-percent improvement could potentially represent approximately $2 million. 

The purpose of this example is not to suggest that every percentage point of improved reliability automatically converts directly into revenue. Manufacturing systems are more complicated than that, and constraints such as market demand, raw materials, labor, process bottlenecks, and product mix must be considered. The larger point is that seemingly small improvements in availability, utilization, and throughput can have meaningful financial consequences when applied across large production systems over an entire year. 

This is the power of marginal gains. Reliability improvement compounds, and the cumulative effect of many small improvements can ultimately be much greater than the value of any single predictive maintenance finding. Remember, change doesn’t happen overnight.  Change happens in small, daily, incremental efforts that work to improve processes that eventually add meaningful value.   

Translating Technical Findings Into Business Decisions 

One of the defining characteristics of a mature predictive maintenance program is the ability to translate technical findings into language that different levels of the organization understand. A reliability engineer may see a developing bearing defect. A maintenance planner sees future corrective work. A maintenance manager sees labor, parts, and schedule requirements. An operations manager sees production risk. A plant manager sees throughput and cost, while a financial leader may ultimately see margin and return on invested capital. Each person is viewing the same physical problem through a different lens. 

An effective ultrasound program should therefore become capable of communicating both the technical condition and its potential business consequence. A statement such as “ultrasound indicated a 16dB increase above baseline” may be technically meaningful to the analyst, but it may have limited meaning to someone responsible for production or financial performance. Communicating that a developing condition was identified on a production-critical motor, validated, and scheduled for repair during an upcoming planned outage provides a much clearer explanation of why the finding matters. 

The technical measurement remains important, but the business context communicates value and the associated risk.  This is how predictive maintenance earns broader organizational support and begins influencing decisions beyond the maintenance department. 

Integrating Technologies Instead of Creating Silos 

Ultrasound as a PdM tool should rarely operate as a completely independent predictive maintenance technology. Different condition monitoring technologies detect different physical characteristics associated with asset degradation, and their combined value can be greater than the value of any one technology used by itself. 

A developing mechanical condition, for example, may create changes in ultrasonic energy before it produces a significant temperature increase. Vibration analysis may then provide additional information regarding the nature or severity of the mechanical condition, while infrared thermography may identify thermal consequences. Oil analysis may reveal contamination or wear debris, and electrical testing may provide additional evidence when an electrical failure mode is suspected. 

The question should therefore not be, “Which predictive maintenance technology is best?” Instead, organizations should ask, “Which combination of technologies gives us the best opportunity to detect this failure mode early enough to take effective action?” Mature reliability programs build technology strategies around assets, failure modes, and consequences rather than around departmental ownership of individual instruments. Ultrasound, vibration analysis, infrared thermography, oil analysis, motor testing, and other technologies should support one another as components of a broader reliability strategy. 

Building a Culture That Sustains the Program 

Technology implementation is relatively easy compared with organizational change. A new instrument can be purchased in a matter of weeks, and an employee can attend a training course in a matter of days. Building a mature reliability culture, however, can take years. 

For an ultrasound program to become sustainable, it cannot depend entirely on one enthusiastic technician, reliability engineer, or program champion. Knowledge must become institutional. Procedures should be documented, routes standardized, baseline and alarm criteria understood, data stored and reported, workflows established, and training continued as personnel and responsibilities change. Management should review meaningful performance indicators, operations should understand why inspections matter, planners should understand how predictive findings enter the backlog, and craft personnel should understand why work generated through condition monitoring may need to be completed before the equipment appears to be broken. 

This last point is particularly important because predictive maintenance intentionally asks organizations to repair equipment that may still appear to be operating normally. That concept can conflict with a deeply reactive culture. Someone may reasonably observe that the machine is “still running,” but mature reliability organizations understand that still running and healthy are not the same thing. The objective is not to see how long an asset can continue operating after degradation begins. The objective is to intervene at the point that minimizes risk, cost, and operational disruption. 

Changing this mindset is an essential part of the reliability maturity journey. 

A Practical Ultrasound Program Maturity Model 

The evolution of an ultrasound program can be viewed across five general levels of maturity. At Level 1, Reactive Use, ultrasound is primarily a troubleshooting tool used after someone already suspects a problem. The technology can provide significant value, but its use is largely event-driven and dependent on individual expertise. 

At Level 2, Developing Program, the organization begins moving toward structured condition monitoring. Routes are created, personnel are trained, critical assets are identified, measurement locations become standardized, and baseline data begins accumulating. The organization is developing repeatability, but the process may still depend heavily on individual program champions. 

At Level 3, Established Condition Monitoring, data is collected consistently and trended over time. Baseline and alarm criteria are developed, findings are documented, inspection intervals become increasingly risk-based, and corrective work is generated from identified conditions. Ultrasound has now become part of the organization’s normal maintenance process rather than primarily a troubleshooting technology. 

At Level 4, Integrated Reliability Program, ultrasound findings are integrated with planning and scheduling, asset criticality, complementary predictive maintenance technologies, root cause analysis, failure and defect elimination, and reliability performance indicators. Information generated by the program is being used not simply to repair assets but to improve maintenance and reliability processes. 

Finally, at Level 5, Business-Driven Reliability, condition monitoring information contributes to decisions involving asset strategy, operational risk, capital investment, asset utilization, throughput, and continuous improvement. Reliability information is no longer confined to the maintenance department. It becomes part of the way the organization manages operational and financial performance.  The ultrasound program is standardized across all sites within the organization and the data is regularly shared between all sites.   

Organizations do not need to reach the highest level immediately, nor will every application mature at the same rate. Sustainable improvement is almost always incremental. The important step is to understand where the program is today, determine what the next level should look like, and consistently close the gap.

Reliability Is the Strategy 

Ultimately, an ultrasound program should not exist simply because ultrasound technology is available. It should exist because an organization has physical assets that must perform required functions and failure modes that can be detected, managed, or prevented through better information. That distinction changes the purpose of the program. 

Ultrasound is not the strategy. Predictive maintenance is not the strategy. The broader strategy is enabling the organization to consistently achieve its operational and business objectives. Reliability supports that strategy by helping ensure that physical assets are capable of performing when they are needed. Predictive maintenance supports reliability by identifying developing conditions before they become functional failures, and ultrasound is one of the tools that enables organizations to accomplish that objective. 

The most mature organizations therefore stop measuring the success of ultrasound solely by how many ultrasonic problems are found. They begin measuring success by what happens because those problems were found. Was a functional failure avoided? Was corrective work planned rather than performed reactively? Was unplanned downtime reduced? Did the asset operate longer? Did availability or throughput improve? Was operational risk reduced? Perhaps most importantly, did the organization learn something from the failure mode that could prevent the same problem from occurring again? 

Those outcomes represent the real value of condition monitoring. The journey may begin with detecting ultrasonic energy from Friction, Impacting, and Turbulence, but the destination is much larger. It is a maintenance organization capable of anticipating rather than reacting, an operations organization capable of planning with greater confidence, and a business capable of extracting greater value and throughput from its physical assets. 

That is what it means to move Beyond the Noise.