Bearing grease analysis is the process of sampling lubricating grease from an operating bearing, then examining chemical composition, wear debris, and contamination indicators to assess equipment condition without dismantling.
This technique belongs to predictive maintenance — similar to vibration monitoring and infrared thermography, but provides different insights: it reveals what is happening inside the bearing housing, not just whether the equipment is vibrating. According to SKF Reliability Systems, grease analysis can detect early-stage failures 2–4 months before vibration-only monitoring alone. This article covers technical fundamentals, analysis methods, result interpretation, and field sampling procedures — based on technical literature from SKF, NSK, NTN, and ISO 281:2007.
Definition and role in predictive maintenance
Bearing lubricating grease performs three simultaneous functions: creates an oil film preventing direct metal-to-metal contact, dissipates heat from the contact zone, and blocks water and dust ingress past the seals. During normal bearing operation, grease gradually accumulates wear products — microscopic metal particles, oxidation fragments, and environmental contaminants.
Grease analysis reads these accumulations like an equipment health record. A sudden spike in iron particle concentration signals accelerated wear. Copper particles suggest cage attack. Water in the grease drives electrochemical corrosion and oil-film breakdown. Each indicator points to a specific damage mechanism — and more importantly, to the root cause that can be remedied.
Grease analysis works most effectively when combined with ISO 10816 vibration monitoring and regular temperature checks. Vibration tells you the severity of the problem; grease analysis tells you the cause and damage type. Together, they enable more accurate maintenance decisions about timing and method of intervention.
| Method | Early detection | Root cause | Sample cost | Notes |
|---|---|---|---|---|
| Vibration monitoring | 2–8 weeks | Partial | ~0 (equipment-based) | Requires baseline |
| Grease analysis | 2–4 months | Complete | $50–200 USD/sample | Requires specialized lab |
| Infrared thermography | 1–3 weeks | Limited | ~0 | Late-stage damage only |
| Acoustic analysis | 1–4 weeks | Limited | ~0 | Requires skilled listening |
Grease analysis justifies its cost most clearly on equipment with high downtime expense or difficult-to-access locations — grinding mills, boiler feed pumps, large industrial fans, bridge crane gearboxes.
Ferrography — wear debris analysis under microscopy
Ferrography is the technique of depositing wear particles from diluted grease onto a glass plate using a gradient magnetic field, then examining under optical and electron microscopy. This method, developed by Westinghouse Research Laboratories in the early 1970s, is now standard diagnostic practice in heavy industrial maintenance.
Two common ferrography types:
Analytical Ferrography (AF): Particles deposit onto a ferrogramme — a glass slide with adhesive coating. A technician reads particle morphology, size, color, and distribution under magnification. Provides the richest qualitative detail but requires experienced interpretation.
Direct Reading Ferrography (DR): Optical measurement of large particles (DL, >5 µm) and small particles (DS, <5 µm). Total Wear Particle Concentration (WPC) = DL + DS; Severity Index (Is) = DL² + DS². When Is doubles or triples across consecutive samples, that signals accelerating damage.
Sample preparation for ferrography requires diluting grease with tetrachloroethylene in ratios from 1:100 to 1:500 depending on base viscosity. NLGI Grade 2 grease typically uses 1:200 ratio. The diluted sample must be shaken evenly and analyzed within 24 hours after dilution to prevent particle settling.
Ferrography works best for bearings at low-to-moderate speeds (<3,000 rpm) under heavy load — such as SRB 22220 EK/C3 (d=100, D=180, B=46 mm, C=365 kN) spherical roller bearings on cement mills. At high speeds, wear particles become too fine and evenly distributed for ferrography to yield actionable data.
Metal particle morphology — cutting, fatigue, adhesive
Metal particle shape in grease is the single most diagnostic indicator. Three main particle types correspond to three distinct damage mechanisms, each requiring different remedial action.
Cutting Wear Particles
Cutting particles are thin ribbons with sharp edges, length 5–10 times width. They form when a harder surface cuts into a softer one — typically when hard wear debris becomes trapped between inner and outer races, or when hard external contaminants enter the grease. High concentrations of cutting particles (>50 particles/mL in diluted solution) combined with particles >15 µm signal external contamination — the dust filter or seals are failing.
Fatigue Wear Particles
Fatigue particles appear as flat flakes or small spheres with smooth surfaces. Flakes emerge when surface fatigue cracks propagate and spall — characteristic of stages 3–4 in the P-F curve bearing life model. Fatigue spheres (1–5 µm diameter) form under localized sliding contact conditions and often appear earlier than flakes. According to NSK Technical Report 2022, a fatigue sphere ratio increase >20% between consecutive samples is a reliable alert threshold.
Adhesive Wear Particles
Adhesive particles are irregular with rough surfaces, often accompanied by heat-discoloration evidence (blue or blue-green particles under microscopy — oxidation from local temperatures >300°C). They form when the oil film ruptures completely and bare metal surfaces contact under load. Common causes: insufficient grease, wrong NLGI grade, or bearing dn (bore diameter mm × rpm) exceeding design limits.
| Particle type | Morphology | Typical size | Root cause | Action |
|---|---|---|---|---|
| Cutting | Thin ribbon, sharp edges | 15–100 µm | Hard contaminant, seal leak | Inspect filter, replace seals |
| Fatigue — flake | Flat, smooth | 20–200 µm | Surface fatigue, late stage | Schedule replacement |
| Fatigue — sphere | Small ball, uniform | 1–5 µm | Early localized sliding | Monitor closely, increase frequency |
| Adhesive | Irregular, rough | 5–50 µm | Oil film rupture | Replenish grease immediately, check load |
| Corrosion | Thin lamella, porous | 1–10 µm | Water, acid, galvanic | Check seals, measure humidity |
Supplementary elemental analysis (spectroscopy) identifies metal composition: iron (races, rollers), copper/brass (cages), chromium (stainless or coatings). The Fe/Cu ratio helps distinguish whether main bearing races or cages are failing — critical for repair decisions.
Contamination — water, dust, and grease incompatibility
Contamination accounts for over 50% of premature bearing failures according to statistics from NTN Industrial Bearing Technical Reference CAT. No. 3017/E. Three contamination types require different detection and remedial approaches.
Water Contamination
Water in grease drives electrochemical corrosion, hydrolysis of thickening agents, and pitting on rolling surfaces. Danger threshold: >0.1% water by weight. Detection methods:
- Dean-Stark test: Distill grease sample, measure condensed water volume — highest accuracy, lab-performed
- Karl Fischer method: Measure water activity via chemical reaction, sensitivity to 10 ppm
- Field rapid test: Place small grease drop on steel plate heated to 150°C — water boils with characteristic pop sound
Common water sources: seal wear, low-temperature condensation (outdoor equipment), cooling system leaks, high-pressure washdown without proper bearing protection.
Dust and Hard Particle Contamination
Hard particles >5 µm scratch rolling surfaces and raceways. Contamination severity follows ISO 4406:2021 — particle count scale for three standard sizes (4 µm, 6 µm, 14 µm). Target for typical industrial bearings: ISO 16/14/11 or cleaner.
Check dust contamination through Particle Count Analysis (PCA): automatic particle count in diluted solution. Comparing results between samples matters more than absolute values — a 30% increase across consecutive samples warrants concern more than any single high reading.
Incompatible Grease Mixing
Mixing two greases with different thickeners — for example lithium complex and polyurea — typically results in gel structure breakdown, grease bleeding at temperatures below design drip point, or hardening. Detected by viscosity measurement (change >20% from baseline) and observing fiber structure under magnification. Grease compatibility charts from manufacturers (SKF, NSK, NTN all provide them) must be consulted before introducing fresh grease into a bearing with residual old grease.
| Contamination indicator | Alert threshold | Danger threshold | Measurement method |
|---|---|---|---|
| Water (% by weight) | 0.05% | 0.1% | Karl Fischer / Dean-Stark |
| Particles ≥5 µm (count/mL) | 5,000 | 20,000 | Automatic particle count |
| Acid number (AN mg KOH/g) | +1.0 from baseline | +2.0 from baseline | ASTM D 974 |
| Iron (ppm) | 50 | 200 | ICP-OES spectroscopy |
| Copper (ppm) | 20 | 80 | ICP-OES spectroscopy |
Color, odor, and consistency — field-side diagnostic tools
Before sending samples to the lab, sensory inspection yields quick diagnostic clues on-site. Experienced maintenance technicians read important signals directly from simple observation.
Color
Fresh NLGI Grade 2 lithium complex grease typically ranges from light yellow to dark yellow or pale green depending on manufacturer. Color changes signal:
- Black or dark gray: High-temperature oxidation or severe iron wear — grease is spent, carbon has formed
- Rusty brown: Electrochemical corrosion, prolonged water exposure
- White or milky opacity: Water contamination >1%, emulsified grease
- Metallic sheen (silver/gold): High metal particle concentration — worrying acceleration of wear
- Pale blue: Extreme local temperature oxidation (>300°C), usually with adhesive particles
Odor
Normal grease smells of mineral or synthetic base oil. Changes in odor signal:
- Burnt, acrid smell: Temperature too high, oxidation or carbon formation
- Sour, acidic smell: Chemical oxidation, water reacting with additives
- Sulfur smell: Extreme Pressure (EP) additive activated — shock load or abnormally high load
- Foreign oil smell: Cross-contamination with hydraulic fluid, engine oil, or cleaning solvent
Consistency and Texture
Pull a small grease sample with a probe:
- Loose, no structure: Thickener breakdown from heat or incompatible mixing
- Hard, chunky: Oxidation, base oil separation, or service life exceeded
- Fine fibers stretching: Good lithium grease, thickener structure intact
- Foamy, porous: Serious water contamination or air entrainment
Record sensory observations on a tracking sheet immediately upon sampling. While not a replacement for lab analysis, sensory data provides essential context for laboratory technicians to interpret results in the correct direction.
Sampling procedure and laboratory analysis
Grease analysis results have value only if the sample represents real bearing conditions. Improper sampling makes every downstream analysis meaningless.
Field sampling procedure (7 steps)
- Preparation: Clean glass jar, chemically inert sealed cap, clean sampling needle. Do not use polyethylene plastic containers — mineral oil soaks into plastic, invalidating viscosity results.
- Timing: Sample after equipment has run for minimum 2 hours — grease is evenly dispersed, wear particles at equilibrium. Sampling immediately after startup yields artificially high particle readings.
- Location: Insert sampling needle into the load zone of the bearing — not just the grease fitting, since that location holds the newest grease and is not representative.
- Quantity: 3–5 grams suffices for complete analysis suite (ferrography + spectroscopy + viscosity + water). Samples under 2 grams lack sufficient material for all tests.
- Label immediately: Equipment name, bearing location, sampling date, hours since last regreasing, cumulative run hours. Missing this information renders lab results uninterpretable in trend context.
- Cool storage: Keep grease at 4–8°C if not analyzed immediately. Do not freeze — destroys gel structure.
- Ship within 48 hours: Metal particles continue reacting within the grease after removal from the bearing.
Standard laboratory analysis package
A complete analysis package for critical equipment includes: analytical ferrography (AF), ICP-OES optical spectroscopy measuring 20+ metal elements, Karl Fischer water content, kinematic viscosity at 40°C and 100°C, and acid/base number (AN/BN) measurement. Turnaround time at industrial labs in Vietnam (Hanoi, Ho Chi Minh City) typically 3–5 business days.
Screening package for lower-priority equipment: elemental spectroscopy + particle count + water content. Cost savings of 40–50% versus full package.
Recommended sampling frequency
| Equipment criticality | Sample interval | Notes |
|---|---|---|
| High (shutdown = line stop) | Every 500–1,000 run hours | Or every 3 months if low-hour equipment |
| Medium criticality (backup unit available) | Every 1,500–2,000 hours | Minimum every 6 months |
| Lower priority | Every 3,000–4,000 hours | Minimum annually |
Establish baseline with 2–3 initial samples from new equipment — this becomes the reference for detecting trends, not for comparing against absolute values from technical literature.
Real-world case — grease analysis prevented blast furnace blower failure
At a steel mill in Binh Duong, a high-pressure turbo blower for the blast furnace runs 24/7 with >99% availability requirement. Main shaft bearing pair: angular contact 7318 BEP (d=90, D=190, B=43 mm, C=118 kN), lubricated with Mobil Polyrex EM polyurea grease.
Maintenance deployed grease analysis every 800 run hours (roughly monthly). The seventh-month sample (normal baseline) looked fine. The eighth-month sample revealed: iron jumped from 18 ppm to 67 ppm (+272%), copper jumped from 4 ppm to 31 ppm (+675%), high concentration of 2–4 µm fatigue spheres appeared, and water measured 0.08% — approaching alert threshold.
Concurrent vibration monitoring (still in Stage 1 per ISO 10816 — no automatic alarm) led technicians to decide on detailed inspection during the next scheduled maintenance window, two weeks out. Upon opening the bearing, they found brass cage showing localized corrosion and the inner raceway exhibiting pitting 0.3 mm diameter at three locations — early-stage fatigue failure, not yet producing obvious noise.
Replacement of both bearings; root-cause investigation: condensation overnight through a worn exhaust valve. Remedy: replace valve gland packing, increase grease schedule from 800 to 600 hours during monsoon months.
Intervention cost: 2 bearing + 4 hours planned downtime. Estimated cost if run to failure: 6–12 hours emergency stop + risk of shaft damage from bearing fracture at full operating speed. Cost-benefit ratio: 15:1 to 30:1 against annual grease analysis program cost.