Setting up an oil analysis programme: lab, on-site tests or route-based
Three ways to run oil analysis, how to choose between them, how to set up a route, and how to count the savings of changing oil by condition.
Three ways to do it
| Approach | How it works | Good for | Watch for |
|---|---|---|---|
| Outside laboratory | Samples are taken, shipped and tested by a lab, with a report and advice | Starting quickly, wide test range, no capital cost | Days of delay; samples labelled badly; sample point and method need discipline |
| On-site laboratory | A small set of instruments in the plant: elemental, particle, viscosity, infrared, water | Many machines, remote sites, quick retest | Needs training, standard procedures and a person to own it |
| Route-based portable tests | A technician walks a route with handheld instruments and tests at the machine | Quick screening of many points, combined with vibration and thermography routes | Needs good records, sample point identification and alarm limits per machine |
Setting up a route
- List each machine and sample point, with the oil type and a reference value for each.
- Fix the sequence of the walk and load the route into the instrument or the sheet.
- At each point, check the reference oil, take the reading and compare with the alarm limits.
- Use green, yellow and red bands: within limits, near the alarm, over the alarm.
- Upload the results and review the trends. Retest on the spot if a reading looks wrong.
- Review the route twice a year: add machines that failed, remove ones that never changed.
Alarm limits and trends
Set warning and alarm values for each property and each oil family. Start with the maker's limits and the new-oil values, then tighten them from your own history. Watch the rate of change as well as the level. A parameter that changes fast needs attention even before it hits the limit.
Counting the savings of condition-based oil change
Compare the cost of a test with the cost of an unnecessary change. As an example, a 35 gallon gearbox change was costed at about 480 US dollars: oil 140, labour for two mechanics for six hours 240, disposal 50 and restock 50. A test costing 15 dollars that proves the oil is still good avoids that cost, a ratio of about 32 to 1. For a plant spending 250,000 dollars a year on lubricants, a 20% reduction by changing on condition is worth about 50,000 dollars in oil, and more once handling, storage and disposal are counted. These are reported example figures; use your own prices.
What a good programme includes
- Testing of all new oil on arrival.
- Organised storage and distribution of lubricants.
- On-site or outside tests for condition and contamination, with trends.
- Portable filtration to return oil to a clean state.
- A named owner, written sample procedures and trained samplers.
Mistakes that make results useless
- Different sample points or methods each time.
- Mislabelled bottles.
- No record of top-ups and changes.
- Alarms never reviewed.
- Testing only when there is already a problem.