An oil analysis report is a laboratory summary of a lubricant sample that quantifies wear metals, additive chemistry, contamination, and physical properties so a maintenance team can judge the health of both the oil and the machine it protects. Read correctly, a single report tells you whether a gearbox is shedding bronze, whether coolant has crept into the sump, and whether the oil still has any protective life left. Read carelessly, it becomes a filing-cabinet formality. This guide walks through the numbers that matter, how alarm limits are actually derived, and a worked example you can apply to your own reports.
Wear metals are reported in parts per million (ppm) by a spectrometer that reads the elemental fingerprint of particles suspended in the oil. Each element points back to a specific family of components, so the pattern matters more than any single number:
One critical limitation: standard emission spectroscopy under-reads particles larger than roughly 5 to 8 microns. A machine failing by fatigue spalling can throw large chunks that the spectrometer barely sees, which is why particle counting and, on critical assets, ferrous density or analytical ferrography are run alongside it. Never declare a gearbox healthy on low iron ppm alone.
Viscosity is the oil's resistance to flow, reported at 40 degrees C (and sometimes 100 degrees C) in centistokes (cSt). It is the property the whole lubrication film depends on, so labs flag even modest drift. A practical rule used by most labs: investigate at roughly plus or minus 10 percent from the new-oil baseline and alarm near plus or minus 20 percent.
Always compare against the correct grade baseline, not against the previous sample. An ISO VG 320 gear oil and an ISO VG 220 will both look normal against themselves and disastrous against each other.
TAN (Total Acid Number) and TBN (Total Base Number) are both reported in mg KOH per gram and describe the oil's chemistry, not the machine's metallurgy. For industrial oils (hydraulics, turbines, gears) you watch TAN: a rising acid number signals oxidation and additive depletion, and corrosion risk climbs as it approaches roughly double the new-oil value. For engine oils you watch TBN: the base reserve neutralizes combustion acids and is consumed over the drain interval. A common condemnation trigger is TBN falling to about half of the new-oil figure, or TAN crossing TBN on the same sample.
Particle contamination is reported as a three-number ISO 4406 code, for example 18/16/13. The three numbers are range codes for particles larger than 4, 6, and 14 microns per millilitre respectively, and each whole-number step is roughly a doubling of particle count. Cleanliness targets tighten with component sensitivity: servo-valve hydraulics might target 16/14/11 while a splash-lubricated gearbox tolerates 19/17/14. Because the codes are logarithmic, a jump from 18/16/13 to 21/19/16 is not a small change, it is roughly an eightfold increase in dirt and demands a filtration and breather check.
Consider three consecutive quarterly samples on the same ISO VG 220 gear oil:
Read the pattern rather than the absolutes. Viscosity is up about 5.5 percent, still inside the investigate band, so the oil itself is only mildly stressed. TAN has risen but is not yet at alarm. The real story is the tight correlation between silicon and iron: silicon has climbed 8x and iron 5x, and the ISO code has jumped four steps at the 14 micron channel. That signature is classic three-body abrasive wear driven by dirt ingress, almost always a failed breather or a leaking seal, not a bearing that is failing on its own. The action is not an oil change, it is to fix the ingress path, flush and filter to restore the cleanliness target, then resample at a shortened interval to confirm iron flattens. This is a textbook condition-based maintenance decision, and catching it here is the difference between a two-hour seal job and an unplanned gearbox rebuild that would wreck your overall equipment effectiveness.
There are two legitimate ways to set limits, and mixing them up causes most false alarms:
The rate of change usually matters more than the level. An iron reading doubling between samples is more alarming than a stable value sitting just under an absolute cap. Pair oil data with reliability metrics such as MTBF and MTTR and, on wear-out-dominated assets, a Weibull analysis to convert a rising trend into a defensible replacement window rather than a guess.
An oil analysis report is only useful if the resulting action lands as a tracked job on the right asset. Fabrico is the real-time data foundation for that loop. As a field-ready CMMS, it holds the asset register, sample history, and spare-parts data, and turns an out-of-limit result into a work order with the correct components and preventive schedule attached. Its CMMS solution keeps every sample, action, and follow-up on the same asset record so trends are not lost in a spreadsheet, while real-time OEE and production monitoring shows the machine-level impact of the downtime you are trying to prevent. Fabrico is EU-built with EU data residency. It does not perform the lab chemistry itself, and predictive maintenance modelling remains an industry practice layered on top of clean, structured maintenance data rather than a finished Fabrico feature. Feeding oil-analysis actions into a structured CMMS is exactly the shift from reactive to proactive maintenance that separates a firefighting shop from a reliable one.
Interval depends on asset criticality, oil volume, and duty. Critical or hard-working assets are often sampled monthly, general industrial gearboxes and hydraulics quarterly, and lightly loaded reservoirs semi-annually. Always sample from the same point, at the same operating temperature, and before top-up, so results are comparable. If a report trends toward an alarm, shorten the interval immediately rather than waiting for the next scheduled pull.
Wear metals such as iron, chromium, and lead come from the machine's own components degrading. Contaminant metals and elements such as silicon (dirt) and sodium or potassium (coolant additives) come from outside the system. Additive metals such as zinc, phosphorus, and calcium come from the oil formulation itself. Reading them together is what separates a genuine wear problem from an ingress or coolant-leak problem.
No. Spectrometry under-reports large particles, so a bearing failing by fatigue spalling can produce alarming ferrography while spectrometric iron still looks modest. Low ppm is reassuring only when particle counts, cleanliness codes, and viscosity all agree. On critical rotating equipment, pair spectrometry with particle counting and periodic ferrography before drawing conclusions.
Ready to turn oil-analysis results into tracked, closed-loop maintenance actions on the right assets? Book a Fabrico demo and see how real-time OEE and a field-ready CMMS keep your reliability data in one EU-hosted place.
Login
Blog
