When Compressors Lie.

Why Normal Operating Conditions Can Hide Serious Internal Problems

By Dr L. Wang
Diagnostics & Condition Monitoring


BEFORE YOU BEGIN READING

One of the most dangerous assumptions in industrial refrigeration is that “everything looks normal.”

The compressor starts on time.

Oil pressure remains within specification.

Discharge temperatures appear acceptable.

No unusual vibration is noticed during routine inspections.

Production continues uninterrupted.

From every visible indicator, the machine appears healthy.

Yet experienced reliability engineers know that catastrophic failures rarely begin with obvious warning signs.

More often, they begin beneath the surface—inside bearings that are slowly deteriorating, lubrication systems that are quietly becoming contaminated, valve assemblies beginning to fatigue, or alignment conditions gradually drifting away from design specifications.

These early changes rarely stop production.

Instead, they become invisible companions to normal operation.

By the time traditional operating parameters finally indicate that something is wrong, the internal damage has often progressed far beyond the point where a simple maintenance intervention would have been sufficient.

This Engineering Paper explores one of the most misunderstood realities of compressor reliability:

Machines can appear perfectly healthy while silently progressing toward failure.

Understanding why this happens is one of the most valuable skills any maintenance engineer can develop.


FROM DR. L. WANG’S DESK

“Throughout my career, I have investigated compressor failures on five continents.

Although every investigation begins with a different machine, industry, or operating environment, one statement appears with astonishing consistency.

“Everything looked normal.”

Those words have accompanied seized bearings, fractured crankshafts, damaged rotors, collapsed valve assemblies and catastrophic lubrication failures.

The operators were not careless.

The maintenance teams were not inexperienced.

The instrumentation wasn’t necessarily inaccurate.

The problem was much simpler.

They were observing the wrong evidence.

Compressors rarely conceal failure.

Instead, they reveal it through signals that traditional operating parameters cannot always detect.

Those signals may appear inside vibration spectra, oil analysis reports, ultrasonic measurements, and motor current signatures.

Or hidden within microscopic wear particles long before temperatures or pressures begin changing.

The purpose of this Engineering Paper is not to encourage unnecessary maintenance.

It is to encourage better observation.

Because machines don’t usually fail without warning.

We simply don’t always recognize the language they’re using.”


ENGINEERING PAPER SNAPSHOT

Industry Focus

Industrial Refrigeration Compressors

Primary Theme

Diagnostics & Condition Monitoring

Applicable Equipment

  • Reciprocating Compressors
  • Screw Compressors
  • Ammonia Refrigeration Systems
  • CO₂ Refrigeration Systems
  • Industrial Process Cooling Plants

Recommended Audience

  • Reliability Engineers
  • Maintenance Engineers
  • Plant Managers
  • Service Engineers
  • Condition Monitoring Specialists
  • Technical Procurement Professionals

Core Engineering Principle

“Normal operating conditions do not always indicate normal internal machine health.”


CHAPTER ONE

The Dangerous Comfort of Normal

Walk through almost any industrial refrigeration plant, and you’ll hear a familiar phrase.

“The compressor seems fine.”

It’s a reassuring statement.

Unfortunately, it is not a diagnostic conclusion.

It is an observation.

There is a significant difference.

Human beings naturally judge machine condition using the senses available to them.

We listen.

We look.

We feel vibration by placing a hand on the machine.

We observe pressure gauges.

We monitor discharge temperatures.

These observations remain valuable.

But they represent only a small fraction of the information a compressor continuously generates.

Imagine visiting a hospital where every patient receives the same medical examination.

A doctor checks body temperature.

Measures blood pressure.

Listens briefly to the heartbeat.

Then declares every patient healthy without conducting blood tests, imaging studies or laboratory analysis.

Most people would immediately recognise the limitations of such an approach.

Yet many industrial compressors are evaluated in exactly this manner.

If the gauges appear normal…

If no abnormal noise is heard…

If production continues…

The machine is assumed to be healthy.

Reliability engineering demands a higher standard.

The absence of visible symptoms should never be mistaken for the absence of developing faults.

Because the earliest stages of mechanical deterioration often occur long before traditional operating parameters begin changing.

That is precisely why catastrophic failures appear so unexpected.

The evidence existed.

It simply wasn’t being observed.

Wangs’s Field Note

“Machines rarely surprise us. They simply reveal their evidence earlier than we choose to investigate.”


CHAPTER TWO

The Difference Between Operating Data and Diagnostic Data

One of the greatest misconceptions in industrial maintenance is the belief that operating data and diagnostic data are interchangeable.

They are not.

Operating data answers one question:

“Is the compressor currently performing its intended function?”

Diagnostic data answers an entirely different question:

“What condition is the compressor in while performing that function?”

A compressor can continue delivering refrigeration capacity while internal wear accelerates.

It can maintain acceptable discharge temperatures while bearing surfaces begin deteriorating.

It can sustain normal oil pressure despite increasing contamination.

Operating performance and mechanical condition often travel together.

But they do not always do so at the same speed.

This distinction forms the foundation of predictive maintenance.

Traditional monitoring tells engineers how the machine is performing today.

Condition monitoring reveals how the machine is likely to perform tomorrow.

Understanding that difference changes maintenance philosophy completely.

Maintenance is no longer scheduled after symptoms become obvious.

It becomes planned while the intervention is still simple, economical and controlled.

That is where modern reliability engineering begins.

Wangs’s Reflection

“Performance tells us what a machine is doing. Diagnostics tells us what the machine is becoming.”


CHAPTER THREE

Silent Failures: When the Evidence Exists but Nobody Looks

Mechanical failures are often described as sudden.

In reality, very few compressor failures are genuinely sudden.

Most are simply undetected.

The distinction is critical.

A connecting rod does not fracture without experiencing fatigue.

A bearing does not seize without lubrication gradually deteriorating.

A discharge valve does not fail without microscopic damage accumulating over thousands of operating cycles.

These processes begin quietly.

Long before production notices.

Long before alarms activate.

Long before temperatures exceed acceptable limits.

The compressor continues operating.

Operators become comfortable.

Maintenance schedules remain unchanged.

The machine develops a reputation for being “reliable.”

Unfortunately, reliability based only on uninterrupted operation can become dangerously misleading.

A compressor may continue producing refrigeration while internal degradation accelerates every hour.

This period—between the beginning of deterioration and the appearance of visible symptoms—is often referred to as the hidden failure interval.

For reliability engineers, this interval represents the greatest opportunity to prevent catastrophic failure.

It is also the period most frequently ignored.

Because nothing appears urgent.

There is no emergency.

No production loss.

No immediate pressure to act.

Yet this quiet interval is precisely where the greatest value of condition monitoring exists.

Finding evidence before failure becomes visible.

Wangs’s Field Note

“Machines don’t fail overnight. They simply spend weeks or months waiting for someone to notice.”


CHAPTER FOUR

Diagnostic Windows Every Reliability Engineer Should Use

Industrial compressors continuously generate information.

The challenge is knowing where to look.

Over many years of field investigations, I have found that nearly every significant compressor failure leaves evidence in one or more of the diagnostic windows.

Understanding these windows transforms maintenance from reactive repair into informed decision-making.

Vibration Analysis

Vibration remains one of the earliest indicators of developing mechanical faults.

Changes in vibration frequency or amplitude often reveal bearing wear, shaft imbalance, coupling misalignment, looseness or valve abnormalities long before operators hear unusual noise.

The objective is not merely to measure vibration.

It is to understand what each vibration signature represents.

A compressor always vibrates.

The question is whether it vibrates normally.

Oil Analysis

Lubricating oil functions as far more than a lubricant.

It is a diagnostic sample continuously circulating through the machine.

Every oil analysis report tells a story.

Microscopic iron particles.

Copper.

Aluminium.

Silicon contamination.

Water ingress.

Oxidation.

Viscosity changes.

These findings frequently reveal developing problems months before mechanical symptoms become apparent.

Ignoring oil analysis is equivalent to ignoring a blood test that already indicates illness.

Thermography

Temperature rarely causes failures.

Instead…

Temperature reflects them.

Thermal imaging allows engineers to identify abnormal heat patterns associated with overloaded bearings, electrical connections, lubrication deficiencies and cooling inefficiencies.

Temperature trends often reveal deterioration more effectively than single temperature readings.

Context matters.

Trend matters.

Comparison matters.

Motor Current Signature Analysis

Electric motors communicate mechanically as well as electrically.

Changes in current signatures may indicate rotor defects, mechanical loading changes or developing compressor problems.

Although frequently overlooked within refrigeration plants, motor current analysis provides another valuable layer of diagnostic confidence.

Ultrasonic Inspection

Many lubrication failures begin long before they become audible.

Ultrasonic instruments detect high-frequency sound generated by friction, leakage and early mechanical distress.

To the human ear, everything sounds perfectly normal.

To ultrasonic equipment…

The warning has already begun.

Bearing Inspection

Bearings are among the earliest components to reveal signs of developing mechanical distress.

Making bearing analysis a cornerstone of effective condition monitoring.

Subtle increases in vibration frequencies, rising operating temperatures, changes in lubrication condition, or the presence of microscopic wear particles in oil samples often indicate bearing deterioration long before audible noise or catastrophic failure occurs.

By monitoring these indicators and analysing long-term trends rather than isolated readings, maintenance teams can identify emerging defects, schedule planned interventions, and prevent secondary damage to shafts, housings, and other critical compressor components.

In reliability engineering, bearing analysis is not merely about detecting wear—it is about understanding the health of the entire rotating assembly before production is ever affected.

Wangs’s Reflection

“One diagnostic tool may suggest a problem. Two tools confirm suspicion. Three independent indicators create confidence.”

The Diagnostic Windows


CHAPTER FIVE

A Compressor That Looked Perfect

Several years ago, I was invited to investigate a reciprocating compressor in a food processing facility.

The maintenance team had no major concerns.

Production remained stable.

Energy consumption had not changed significantly.

Operators described the compressor as dependable.

Routine operating parameters supported their confidence.

Discharge temperatures remained within specifications.

Oil pressure appeared healthy.

There were no significant complaints from operations.

Yet one detail attracted my attention.

A recent oil analysis showed a gradual increase in iron wear particles.

Not alarming.

Just unusual.

A follow-up vibration analysis identified a subtle increase in bearing-related frequencies.

Again…

Nothing severe.

Individually, neither finding justified shutting down the compressor.

Together, however, they suggested a developing bearing problem.

Management agreed to inspect the machine during the next planned maintenance shutdown.

The inspection confirmed early-stage bearing deterioration caused by lubrication contamination.

The bearing had not yet failed.

The crankshaft remained undamaged.

Production experienced no interruption.

The repair required only a planned bearing replacement and lubrication system cleaning.

The maintenance manager later remarked that, had the compressor continued operating for another two or three months, the repair would almost certainly have involved crankshaft damage, extended downtime and significantly higher costs.

Nothing about the compressor appeared abnormal.

Everything about the diagnostics did.

That experience reinforced an important lesson.

Operating data described the compressor’s performance.

Diagnostic evidence revealed its future.


K9 Reliability Insight

The most valuable repair is the one completed before production ever realises it was necessary.


END OF PART I

PART II WILL CONTAIN:

  • The K9 Decision Point™ for predictive diagnostics
  • Why confirmation bias causes engineers to miss failures
  • A Practical Condition Monitoring Checklist
  • Frequently Asked Questions
  • Conclusions
  • Dr L. Wang’s closing message to maintenance professionals

About The K9 Reliability Series

The K9 Reliability Series is an educational initiative by K9 Spares, created to advance practical engineering knowledge within the industrial refrigeration industry.

Rather than focusing on products, each Engineering Paper explores the principles, maintenance strategies and reliability practices that help engineers improve equipment performance, reduce downtime and make informed maintenance decisions.

Every paper is guided by one philosophy.

Knowledge Engineered. Reliability Delivered.
Built from Experience. Shared with Purpose.


If this Engineering Paper helped you think differently about compressor reliability…

…then it has achieved its purpose.

When the time comes to source dependable OEM-grade compressor spare parts compatible with Mycom, Sabroe, Grasso, Vilter, Frick, York, Carrier, Bitzer and other leading industrial refrigeration compressors, the K-nine Spares team will be ready to help.

Until then…

Keep listening to your machines.

They usually tell you everything you need to know.


K9 Reliability Series

Contributors

David K. Shenoy – Reliability Engineering
Evan Stratton – Compressor Systems
Michael Thornton – Asset Management
Sabah Connoley – Energy Efficiency
Dr L. Wang – Diagnostics
James O’Connor – Plant Maintenance


You may also want to read
✅ Engineering Paper 001

Using Vibration Analysis to Predict Compressor Component Failures

✅ Engineering Paper 002

The Silent Killers of Industrial Refrigeration Compressors