How to plan a gear oil switch trial and its sampling interval
Summary
A gear oil switch trial takes five steps: baseline the old oil with two or three samples, sample the new fill, then fix interval, tests and stop limits. Sample more often than the routine interval, take the first trial sample within one to two weeks (illustrative timing), and judge every sample against that gearbox's own baseline.
ASTM D5185 determines 22 elements, covering wear metals, additive elements and contaminants, in used and unused oils, and wear metal concentrations can indicate abnormal wear if baseline data exist for comparison (ASTM D5185, 2026).
One gear unit maker lists an oil consistency check every 3,000 operating hours, at least every 6 months, and recommends a laboratory analysis when the oil condition is unclear (SEW-Eurodrive, 2026).
ISO viscosity grades allow a tolerance of 10% either side of the grade at 40 °C, so an ISO VG 460 oil may measure 414 to 506 cSt when new (Design World, 2018).
ASTM D6224 covers sampling and testing schedules for mineral gear oils in auxiliary power plant equipment and does not address most synthetic lubricants (ASTM D6224, 2023).
This plan fits enclosed industrial gearboxes with an oil sump or a circulating oil system. It does not cover grease-lubricated gearboxes, automotive axles or wind turbine gearboxes, which follow their own maker and application standards.
What do you need before a gear oil switch trial starts?
Four things must be in place before a gear oil switch trial starts: the gearbox maker's approval of the new oil, a written trial plan, one fixed sampling point, and a laboratory that runs the same methods on every sample. The nameplate and the operating instructions give the approved oil type and viscosity grade. ANSI/AGMA 9005-F16 gives guidelines for minimum performance characteristics of industrial gear oils and states that it is not intended to supplant the specific recommendations of gear manufacturers (ANSI/AGMA 9005-F16, 2016).
The plan names five items in advance: the gearboxes in the trial, the old-oil baseline samples, the new-oil fill sample, the sampling interval and tests, and the stop limits. It also records whether the switch crosses base oil types, because that decides whether the gearbox needs a flush. One gear unit maker does not permit mixing synthetic oils with mineral oils or other synthetic oils, and requires a thorough flush with the new oil grade when switching from mineral to synthetic oil or between synthetic bases (SEW-Eurodrive, 2026).
Read the safety data sheet (SDS) of both oils before the drain and fill, and handle drained oil under your site's used-oil rules. Oil and housing are hot after running: one maker tells users to let the gear unit cool below 50 °C before working on it (SEW-Eurodrive, 2026).
Why do you need baseline samples of the current gear oil before switching?
Baseline samples of the current gear oil matter because every later result in the trial is read as a change, and a change needs a starting point measured on the same gearbox. ASTM D5185 states that wear metal concentrations "can be indicative of abnormal wear if there are baseline concentration data for comparison" (ASTM D5185, 2026). Without a baseline, a normal iron level for a heavily loaded gearbox can look like a failure, and a real rise can look normal.
Take two or three samples of the old oil from the trial sampling point in the weeks before the switch. One reading tells you little; several show the oil's normal spread. Record operating hours, load and oil temperature with each sample. Also keep a sealed sample of the unused new oil from the same batch: the Bureau of Reclamation advises keeping a sample of new, unused oil of the same type and manufacturer for comparison with used samples (Bureau of Reclamation, 2024).
The old-oil baseline also predicts how the new oil's first readings will look, because some old oil stays in the housing, coolers and pipes after the drain. If 10% of the old charge remains and the old oil carried 300 mg/kg of an additive element the new oil lacks, the first sample will read about 30 mg/kg of that element (illustrative figures). The first readings are a blend of the two oils.
How often should you sample the gearbox during a gear oil trial?
Sample the gearbox more often during a gear oil trial than in routine monitoring, and write the trial interval into the plan before the new oil goes in. Routine intervals are the reference point: one gear unit maker lists an oil consistency check every 3,000 operating hours, at least every 6 months (SEW-Eurodrive, 2026). A trial needs data sooner, because a compatibility problem or a wrong fill shows up in the first days of running.
A workable trial schedule looks like this, with illustrative timing:
Fill sample: after the new oil has circulated at operating temperature for the first hours of running.
First trial sample: within the first one to two weeks of running.
Trend samples: every two to four weeks until the trial ends.
Routine interval: back to the maker's interval once the trial is judged.
Take every sample from the same point, either from a fixed sample tube or valve while the gearbox runs (Bureau of Reclamation, 2024), or after the gear unit has cooled below the maker's limit and been run briefly to mix the oil (SEW-Eurodrive, 2026). The Bureau of Reclamation advises flushing a fixed sample tube with at least five times its dead volume first, and testing more often when a sudden increase in contaminants appears (Bureau of Reclamation, 2024).
Which oil analysis tests track a gear oil switch trial?
Five oil analysis tests track a gear oil switch trial: viscosity, water, acid number, wear metals and additive elements. Run them by the same methods on the baseline, fill and trial samples. The Bureau of Reclamation lists viscosity, water content, total acid number, particle count and elemental analysis for wear metals and additives as periodic laboratory tests (Bureau of Reclamation, 2024).
Viscosity: kinematic viscosity at 40 °C by ASTM D445, the method for transparent and opaque liquids (ASTM D445, 2026).
Water: coulometric Karl Fischer titration by ASTM D6304, which reaches from mg/kg to percent levels with a suitable sample size (ASTM D6304, 2025).
Acid number: potentiometric titration by ASTM D664. Its scope notes that additives such as inhibitors and detergents can show acidic characteristics in new and used oils (ASTM D664, 2024), so the new oil has its own starting value.
Wear metals and additive elements: ICP-AES by ASTM D5185, which reads iron, copper and additive elements in one run (ASTM D5185, 2026).
Ask the laboratory to report the method number with each result. Switching laboratory or method in the middle of a trial breaks the comparison, so name both in the plan. Add particle count where the gearbox has fine filtration or the maker sets a cleanliness target.
How do you check gear oil trial results against the baseline?
Check gear oil trial results by comparing each reading with one of two references: the new-oil fill sample for oil condition, and the old-oil baseline for wear. Viscosity, water and acid number describe the new oil, so they are read against the fill sample. Iron and copper describe the gearbox, so they are read against the wear level the old oil showed in that same gearbox.
When a reading breaks the pattern, confirm it with a resample from the same point before acting, because one bottle can carry sampling error. Record operating hours, load and oil temperature with every result.
For each trial sample, the check table below decides whether to continue, resample, or stop and revert.
Check | Pass | Fail and what to do |
|---|---|---|
Viscosity at 40 °C (ASTM D445) | Inside the grade band of 10% either side of the ISO VG grade (Design World, 2018) and close to the fill sample | Outside the band or drifting: resample, check top-ups and residual old oil of another grade, stop if confirmed |
Water (ASTM D6304) | Near the fill sample and under the maker's or the plan's limit | Rising: check breather, seals and cooler, resample within days |
Acid number (ASTM D664) | Steady against the fill sample; the rate of increase matters most (Bureau of Reclamation, 2024) | Sudden rise: contact the oil supplier, apply the plan's stop limit |
Iron and copper (ASTM D5185) | Inside the spread of the old-oil baseline | Rising above the baseline spread: resample, add particle or vibration checks, stop at the plan limit |
Additive elements (ASTM D5185) | Moving from the blended first reading toward the fill sample | Stuck at a blend or moving away: suspect wrong oil or old-oil top-up (ASTM D5185, 2026) |
The additive row settles most early questions. ASTM D5185 notes that additive concentrations far from their specification "can indicate that the incorrect oil is being used" (ASTM D5185, 2026), so a first sample that sits between the two oils points to residual old oil, not a failing new oil.
When should you stop a gear oil switch trial?
Stop a gear oil switch trial when a confirmed result crosses a stop limit written into the plan before the new oil went in, or when the gearbox shows a fault the oil could explain. A result counts as confirmed when a resample from the same point gives the same answer. Typical stop limits, set per gearbox and illustrative here, are iron rising above the old-oil baseline spread, viscosity leaving its grade band, a sudden acid number rise, foaming or deposits, and a rising sump temperature at the same load.
Write the limits as numbers and actions, and do not move them once results arrive. A guide on setting success and stop criteria for a chemical trial covers how to fix them in advance. Where the gearbox has enough history, ASTM D7720 describes statistical techniques to evaluate alarm limits from a unit's own oil analysis data, and states that these techniques do not supersede limits established and validated by the equipment maker (ASTM D7720, 2025).
Plan the revert too: which oil goes back in, who approves the drain, and whether the gearbox needs a flush. Bureau of Reclamation guidance recommends that a system converted to a different lubricant be cleaned and thoroughly flushed before the new lubricant is added (Bureau of Reclamation, 2024).
When should you call the gearbox maker or a lubrication specialist about a gear oil trial?
Call the gearbox maker or a lubrication specialist about a gear oil trial at two points: before the fill when the switch crosses base oil types, and during the trial when a stop limit is crossed or a result is unclear. The first group also takes in gearboxes under warranty, gearboxes with circulating systems and coolers, and switches to an oil the nameplate does not list. The second takes in results that disagree with each other, such as rising iron with steady particle counts.
Give the specialist three things: the plan as written before the trial, the baseline and fill results, and every trial result with its operating hours, load and temperature. A specialist can judge a result only against the conditions it was taken under. The maker can also confirm whether its own limits replace the plan's; where the maker's limits exist, they take priority.
Lubinpla, an applied AI company for specialty chemical technical work, prepares the trial plan from your own documents, so the analysis items and sample plan are fixed before results arrive and each sample carries its transport and handover record. Keep the plan, the chain of samples and the decisions in one file, so a second reviewer can follow how the trial reached its result.
Frequently asked questions
Can you top up with the new gear oil instead of draining the old oil?
Topping up mixes two products and makes the trial unreadable: no result can be assigned to either oil. One gear unit maker does not permit mixing oils of different grades or manufacturers (SEW-Eurodrive, 2026). Drain fully, flush where the maker requires it, and fill with the trial oil.
How long should a gear oil switch trial run?
Run the trial until the trend samples are stable against the baseline across several intervals, and at least long enough to cover the load and temperature range the gearbox sees in normal service. A seasonal plant may need a trial that spans both a warm and a cold period. Write the duration into the plan.
Does a gear oil trial need a control gearbox?
A control gearbox helps when two or more identical gearboxes run under similar load: one keeps the old oil and serves as a running baseline. Without one, a change in production load during the trial can look like an oil effect, so record load with every sample.