Mechanical Monitoring

Compressor Monitoring

Compressor monitoring applies vibration analysis, process parameter monitoring, and oil analysis to centrifugal and reciprocating compressors to detect rotor instability, valve failure, bearing degradation, and performance deterioration.

Core Principle

Centrifugal and reciprocating compressors have fundamentally different failure modes and require different monitoring approaches. A centrifugal compressor can surge into instability within seconds; a reciprocating compressor degrades over weeks through valve wear. Monitoring must be matched to the machine type and its specific failure mechanisms.

Centrifugal Compressor Monitoring

Centrifugal compressors, including blowers and turbochargers, run at high speeds on fluid film bearings. The fluid film bearing design means that contact-based wear is not the primary failure mode under normal operation. Instead, the dominant risks are aerodynamic instability, rotor dynamics, and bearing film breakdown. Each has a distinct monitoring signature.

Rotor Instability and Surge

Surge occurs when the compressor operates too far left on its performance curve, causing the gas flow to momentarily reverse. In a single surge event, pressure swings rapidly between the discharge and suction sides. Repeated surge cycles are destructive to seals, blades, and bearings: the impeller experiences alternating thrust and radial loads it was not designed to sustain continuously.

Process monitoring tracks suction and discharge pressure alongside mass flow. As operating conditions push toward the surge line, an anti-surge control system opens a recycle or blowoff valve to increase flow through the machine and move the operating point back into the stable region. Vibration monitoring during an actual surge event shows characteristic high-amplitude, low-frequency oscillation that is clearly distinguishable from normal running signatures.

Sub-Synchronous Vibration

Fluid film bearings can develop oil whirl instability at approximately 0.42x to 0.48x running speed. This is a self-excited vibration driven by the oil film dynamics in the bearing. If the whirl frequency coincides with the rotor's first critical speed, it can lock there as oil whip, a more severe and potentially damaging condition.

API 670 (Machinery Protection Systems) specifies proximity probe monitoring requirements for large centrifugal compressors: measurement locations on the shaft at each bearing, required transducer pairs oriented 90 degrees apart, and the calculation method for alarm and trip levels relative to the API-allowable vibration amplitude. The standard exists specifically because sub-synchronous events can progress to bearing failure faster than route-based monitoring would detect them.

Impeller Rub

Contact between the rotating impeller and a stationary seal or labyrinth produces a distinctive vibration signature: impulses in the time waveform, possible changes in 1x amplitude and phase as the rub affects rotor balance, and a temperature rise at the rub location if instrumentation is present there. Impeller rub can escalate quickly from intermittent contact to sustained contact, which will cause seal damage and potentially impeller damage.

Early detection relies on monitoring for the characteristic impulses in the vibration signal and any 1x phase shift that suggests altered rotor dynamics. Process-side indicators, such as a reduction in compression efficiency or a change in seal leakoff rate, provide supporting evidence.

Bearing Degradation

On fluid film bearings, bearing temperature and oil analysis are the primary monitoring methods. A rising bearing temperature indicates that the oil film is thinning, whether from high load, reduced oil supply, or contamination. Oil analysis detects wear metals that indicate the film has broken down and metal-to-metal contact is occurring. Vibration provides secondary confirmation as damage progresses, but by the time vibration is clearly elevated on a fluid film bearing, the damage is already significant.

Failure Mode Primary Monitoring Method Key Indicator
Surge Process parameter monitoring Flow approaching surge line; high-amplitude low-frequency vibration during event
Sub-synchronous vibration (oil whirl/whip) Proximity probes (API 670) 0.42x to 0.48x running speed vibration component
Impeller rub Vibration monitoring Impulses in time waveform; 1x amplitude and phase change
Fluid film bearing degradation Bearing temperature; oil analysis Rising bearing temperature; wear metals in oil

Reciprocating Compressor Monitoring

Reciprocating compressors are piston-cylinder machines where each piston stroke produces a pressure pulse and a mechanical impact. The vibration signature of a healthy reciprocating compressor is inherently complex: multiple impacts per revolution, harmonics of crankshaft frequency extending into the kilohertz range, and cylinder-to-cylinder variation. The diagnostic approach is fundamentally different from rotating machinery monitoring.

Valve Condition Monitoring

Cylinder valves are the highest-wear component in a reciprocating compressor and the most common source of unplanned downtime. Each cylinder has separate suction and discharge valves, and any individual valve can degrade while others remain healthy.

Cylinder pressure analysis, using pressure transducers mounted in each cylinder, captures the full compression cycle as a pressure-volume (p-V) diagram. The shape of this diagram encodes valve condition. A leaking discharge valve shows characteristic rounding of the discharge knee: instead of the sharp break where compression work ends and gas enters the discharge line, the curve rounds off as gas leaks back through the damaged valve. A leaking suction valve affects the shape of the suction phase. Valve performance analysis can identify which specific valve on which cylinder is degrading before the efficiency loss becomes operationally significant.

Vibration-based valve monitoring is an alternative approach using accelerometers mounted on or near each valve cover. Valve opening and closing events produce characteristic impulses; changes in timing, amplitude, or the presence of additional impacts indicate valve degradation. This method does not require cylinder pressure taps and is therefore easier to retrofit.

Piston and Rod Condition

Piston rod drop monitoring uses proximity probes to measure the vertical position of the piston rod relative to the stuffing box. As rider bands (the wear pads on the piston that maintain clearance from the cylinder bore) wear down, the piston drops. The rod then contacts the packing rings in the stuffing box, causing accelerated packing wear and gas leakage past the seal.

Rod drop monitoring provides direct, continuous measurement of rider band wear state. It gives maintenance teams advance notice before the piston contacts the cylinder bore, which would require a full cylinder rebuild rather than a rider band replacement.

Frame Vibration

Reciprocating compressor frames generate vibration at integer multiples of crankshaft frequency: 1x, 2x, 3x, and higher harmonics. The specific pattern reflects the number of cylinders and their firing order. Elevated overall frame vibration, or a change in the amplitude of specific harmonics, indicates loose foundation bolts, worn main crankshaft bearings, or connecting rod bearing wear.

Frame vibration monitoring uses accelerometers or velocity sensors at specified locations on the frame, crosshead guide, and main bearing locations. Baseline measurements taken during commissioning or after maintenance provide the reference for change detection.

Cylinder Temperature Monitoring

Each cylinder's discharge temperature reflects the compression work performed and the condition of the valves. A cylinder with a leaking discharge valve will show a higher discharge temperature than the others: gas that leaks back has already been compressed, so the cylinder must re-compress it, increasing the work input. A significantly elevated discharge temperature on one cylinder relative to others is a reliable indicator of valve trouble or internal bypass, even before the p-V diagram analysis confirms which valve is the cause.

Failure Mode Primary Monitoring Method Key Indicator
Valve leakage (discharge) Cylinder pressure analysis (p-V diagram) Rounded discharge knee in p-V curve
Valve leakage (suction) Cylinder pressure analysis Altered suction phase shape in p-V curve
Rider band wear (piston drop) Rod drop monitoring (proximity probe) Increasing rod drop measurement over time
Main/connecting rod bearing wear Frame vibration monitoring Elevated or changing harmonic amplitudes at crankshaft multiples
Internal bypass or valve fault Cylinder discharge temperature One cylinder significantly hotter than others

Screw Compressor Monitoring

Screw compressors are twin-rotor positive displacement machines where a male rotor and female rotor mesh to trap and compress gas. They bridge the monitoring requirements of centrifugal and reciprocating compressors: they run at moderate to high speeds, produce characteristic mesh frequency vibration, and rely on oil injection for sealing and cooling.

Vibration Monitoring

The dominant vibration frequency in a screw compressor is the rotor mesh frequency: the number of lobes on the rotor multiplied by the rotational speed in Hz. The male rotor (typically four or five lobes) and female rotor (typically six lobes) each produce their own mesh frequency and harmonics. A healthy screw compressor shows a predictable spectrum dominated by these mesh frequencies and their harmonics.

Changes in this spectrum indicate specific developing faults. Elevated harmonics of the mesh frequency suggest rotor tip clearance changes. A timing gear fault produces sidebands around the mesh frequency or elevated timing gear tooth-pass frequency. Rotor imbalance shows up as elevated 1x running speed amplitude. Deteriorating rotor lobe coating, which is common in oil-injected screw compressors handling wet or corrosive gas, changes the mesh frequency amplitude progressively over time.

Oil Analysis

In an oil-injected screw compressor, the oil is injected directly into the compression chamber to seal the rotor clearances and carry away compression heat. This means the oil is exposed to the full discharge temperature and to any contaminants in the gas stream. Oil condition analysis tracks oxidation products (which indicate thermal stress), particulate contamination (which indicates rotor wear or external ingestion), and water content (which indicates process gas contamination or condensation).

Oil analysis intervals for screw compressors are typically more frequent than for external oil systems because the oil serves as both lubricant and process sealing fluid, and its degradation directly affects compression efficiency.

Discharge Temperature

Excessive discharge temperature is the primary early indicator of problems in oil-injected screw compressors. Normal discharge temperatures depend on the pressure ratio and gas properties, but most oil-injected screw compressors operate between 80 and 100 degrees Celsius at the discharge. Temperatures above the normal range indicate oil starvation (reduced oil injection rate), an elevated pressure ratio (possibly from a partially blocked discharge or a downstream restriction), or internal rotor-to-rotor contact. A temperature rise of 15 to 20 degrees above the established baseline warrants investigation before it becomes a shutdown event.

Performance Monitoring Across Compressor Types

Efficiency trending detects developing problems across all compressor types before any fault signature appears clearly in vibration data. A compressor that is slowly fouling, wearing internally, or losing valve sealing will show a gradual decline in thermodynamic efficiency over weeks or months. This decline often precedes any threshold-crossing vibration or temperature alarm by a significant margin.

The efficiency metric differs by compressor type:

Tracking these metrics continuously, rather than calculating them only during planned inspections, allows trending that would otherwise be invisible. A centrifugal compressor that loses two percentage points of polytropic efficiency over six months is degrading, but no single measurement would flag it. Continuous trending makes the gradient visible.

Performance monitoring requires accurate flow, pressure, and temperature instrumentation with calibrated sensors. Instrument drift in any one measurement corrupts the calculated efficiency and produces false trends. Periodic sensor validation is part of the performance monitoring program, not separate from it.

API 670 and Monitoring Standards

API 670 (Machinery Protection Systems) is the primary industry standard governing proximity probe systems on rotating machinery. For centrifugal compressors, API 670 specifies the number and placement of proximity probes at each bearing, the orientation (X-Y probe pairs at 90 degrees), the required instrumentation performance specifications, and the method for setting alarm and trip levels relative to the calculated API-allowable vibration amplitude for the machine's bearing clearance and running speed.

The standard distinguishes between protection (tripping the machine before damage occurs) and monitoring (recording data for diagnostic purposes). Both functions use the same instrumentation but serve different purposes: protection systems must meet stringent reliability requirements because a failure to trip when required can allow a catastrophic failure.

API 670 applies to machines where a vibration trip can prevent a catastrophic failure: large centrifugal compressors, gas turbines, large pumps, and similar critical rotating equipment. The standard does not apply to small rotating equipment where the cost of a full API 670 system exceeds the value of the machine.

Reciprocating compressor protection is addressed separately under API 618 (Reciprocating Compressors for Petroleum, Chemical, and Gas Industry Services). API 618 covers design, materials, and instrumentation requirements including cylinder pressure monitoring, rod drop monitoring, and vibration monitoring for reciprocating machines in critical service.

Standard Scope Key Requirements
API 670 Rotating machinery protection systems Proximity probe placement and orientation; alarm and trip level calculations; instrument specifications
API 618 Reciprocating compressors Cylinder pressure monitoring; rod drop monitoring; frame vibration limits
ISO 13709 Centrifugal pumps (referenced by compressor standards) Shaft vibration limits by bearing clearance and speed

Continuous Compressor Monitoring with Tractian

Tractian's sensors monitor centrifugal and screw compressor bearing vibration and temperature continuously, tracking the indicators of developing faults before they affect process performance.

See Mechanical Monitoring
mechanicalmonitoring.com is an educational reference resource. Content describes general engineering principles and monitoring methodologies. Specific alarm levels, trip setpoints, and inspection intervals must be determined by qualified engineers based on machine design, operating conditions, and applicable standards. Always consult OEM documentation and relevant API standards before implementing a monitoring program.