Lubricant Failure Case Record Template for Technical Service

What is a lubricant failure case record template for, and who fills it in?

A lubricant failure case record template is a fixed form for one failure of a lubricated component, such as a bearing, gear set or hydraulic pump, that a supplier's technical service engineer, a plant reliability engineer or a distributor's application engineer fills in when the failure is reported and again when the fix has been running for a while. Its purpose is comparison: two cases recorded on the same fields can be laid side by side, while two cases written as free-text stories cannot.

Use the record when a customer reports a failure that may involve the lubricant, when a used-oil sample comes back with an abnormal result, or when a plant wants a history of failures it can search before the next one. One record covers one failure event on one component. A second failure on the same component, even weeks later, gets its own record with a link to the first, so that recurrence is visible instead of being folded into an old file.

The field names borrow from reliability data practice. ISO 14224:2016, which sets reliability and maintenance data collection for equipment in the petroleum, petrochemical and natural gas industries, describes its failure modes as usable as a "reliability thesaurus" (ISO 14224:2016). Plants outside those industries can borrow the same vocabulary without claiming conformity. The editable Word record, an Excel register and a CSV of the register columns are at the end of this page.

Which sections and fields does the lubricant failure case record template have?

The lubricant failure case record template has eight sections, and every field in them has a fixed name, a unit and a rule that an unknown value stays blank. The table below previews the fields of every section and decides whether two cases can be compared on the same axes; the example values in the last column are illustrative and come from no real site.

The key rows are the failure mode term and the cause with its certainty, because a search for similar cases matches on those two fields first.

In the Excel register, each case is one row and each field above takes one or more columns, so a filter on equipment type, failure mode term or fluid grade shows every comparable case at once. Units sit in the column headers rather than in the cells, which keeps the values sortable. A field that does not apply, such as the ISO 4406 code for a grease, is written as "n/a"; a field that applies but was not measured stays empty. Keep the field names fixed across sites and years, because renaming a field splits one history into two.

Which equipment and operating conditions should a lubricant failure record capture?

A lubricant failure record should capture the equipment class and tag, the failed component, the materials the fluid touches, and the operating ranges of oil temperature, load and speed, each with its unit and where it was measured. These fields decide whether a past case is similar at all: two bearing failures on the same grease are not comparable if one ran at 60 °C and the other at 110 °C (illustrative).

The ISO 14224 failure record, as presented by its technical editor in 2014, lists equipment identification and location, the subunit failed, the component or maintainable item failed, the detection method, and the operating condition at failure, such as running, start-up, testing, idle or standby (Selvik, 2014). The template keeps those fields and adds the wetted materials, because seal elastomers, yellow metals and coatings can react with a lubricant's additives and change which cause is plausible.

Record operating conditions as ranges with units, not as words. "Hot" or "heavy load" cannot be compared; "65 to 78 °C at the sump" can (illustrative). Write where each value came from: a plant historian, a handheld reading on the day, or the operator's estimate. An estimate is marked as an estimate. A value nobody measured stays blank rather than being filled with the design figure, because a design figure recorded as an operating value makes two different cases look alike.

Lubricant failure record fields for the fluid: product, grade and service history

The fluid section of a lubricant failure record names the product as it appears on its data sheet, the viscosity grade, the base oil type, the components the documents disclose, the hours since the last oil change, and any top-ups or mixing with another product since then. A fluid recorded only as "gear oil" cannot be matched with anything.

Known components come only from the technical data sheet, the safety data sheet or a written statement from the maker. If the TDS says the oil carries an extreme pressure additive of a sulfur-phosphorus type, record that and name the document and its revision. If the documents do not disclose a component, write "not disclosed" rather than guessing from the product family. A guessed additive in a record becomes a false match later, when someone searches for failures with that chemistry.

Service history is where many failure records go thin. Record the hours in service in hours, the date of the last oil change, the volume and product of each top-up, and whether a different product was added. Mixing two lubricants is a change of fluid, so the record names both products and the approximate share of each if it is known. If the history is unknown, the field says "history unknown", which is itself a finding the next engineer needs to see.

How do you record lubricant failure evidence such as oil analysis results and photos?

Record lubricant failure evidence as values with units and the test method behind each one, next to the new-oil value or the last normal sample, and keep photos and lab reports as named attachments linked by file name and report number. A used-oil value alone says little; the change from the baseline is what a later case can be compared on.

Use the unit each test method reports. Kinematic viscosity under ASTM D445-26 is reported in mm2/s, where 1 mm2/s equals 1 cSt (ASTM D445-26). Water by coulometric Karl Fischer under ASTM D6304-25 is stated in mg/kg, with a precision range of 20 mg/kg to 25 000 mg/kg for Procedure A (ASTM D6304-25). Acid number under ASTM D664-24 is stated in mg KOH/g, with 0.5 mg KOH/g to 90 mg KOH/g in the precision statement for Test Method A (ASTM D664-24). Particle contamination under ISO 4406:2021 is a code of three scale numbers for particles at or above 4 µm(c), 6 µm(c) and 14 µm(c) per millilitre (ISO 4406:2021, clause 4.2).

ASTM D7669-20 is a guide to trend analysis of lubricant condition data, and it states that it does not address upper or lower control limits, which come from product manufacturers or ASTM specifications (ASTM D7669-20). The record therefore names the source of any limit it compares a value against. Photos carry the date, the component and what the photo shows, so a reader can judge the failure mode without the original file open.

How do you mark the cause of a lubricant failure as confirmed or inferred?

Mark the cause of a lubricant failure as confirmed when a check has ruled out the other candidates, as inferred when the evidence points to it but has not excluded the rest, and as unconfirmed when it is a hypothesis nobody has tested. Write the certainty in its own field next to the cause, together with the evidence that supports it.

Keep the failure mode, the failure mechanism and the failure cause apart. In the ISO 14224 failure record, failure mode is minimum data that shall be collected, while the failure mechanism is "the physical, chemical or other processes which have led to a failure" and the failure cause is the circumstances during design, manufacture or use behind it (Selvik, 2014). Corrosion on a bearing raceway is a mode; water ingress through a worn breather is a cause (illustrative). The ISO 14224 tables group causes into design-related, fabrication or installation-related, operation or maintenance, management, and miscellaneous classes (Selvik, 2014), which gives the cause field a short controlled list.

For rolling bearings, ISO 15243:2017 classifies in-service failure modes for bearings made of standard bearing steels by their appearance (ISO 15243:2017). SKF describes its six categories as rolling contact fatigue, wear, corrosion, electrical erosion, plastic deformation, and cracking and fracture (SKF Evolution). The standard notes that if visual examination cannot reliably assess the root cause, additional investigations are to be considered (ISO 15243:2017), which is the point at which a cause stays inferred.

Lubricant failure record fields for action and outcome

The action and outcome fields of a lubricant failure record state what was changed, who changed it and when, and then how long the fix held, measured in running hours or days until the failure came back or until the review date. A record that ends at the action cannot tell the next engineer whether that action worked.

Write each action as one change: an oil change to the same product, a change of product, a new breather, a filtration change, a seal replacement or a change of relubrication interval. If two changes were made at once, record both and note that the outcome cannot be credited to either one alone. This matters for later comparison, because a case where three things changed together is weaker evidence for any one of them than a case where one thing changed.

Set the review date when the action is recorded, and fill in the outcome on that date even if nothing happened. "No recurrence after 2,900 h, reviewed 2027-03-02" is an outcome (illustrative); a blank outcome field is a case that nobody followed up. If the component failed again, open a new record, link the two case IDs in both records, and change the earlier record's recurrence field to yes. The earlier cause then drops back to inferred unless the second failure confirmed it.

How to fill in the failure case record, with an illustrative example

Fill in the failure case record in two visits: sections 1 to 6 when the failure is reported, and sections 7 and 8 when the action is taken and at the review date. Write the cause field after the evidence, not before it.

1. Open a case ID and record the date detected and the anonymized site.

2. Record the equipment, component and wetted materials from the asset register or the nameplate.

3. Copy the fluid name, grade and disclosed components from the TDS and SDS, with their revisions, and record the service history.

4. Record operating ranges with units and where each came from; leave unmeasured values blank.

5. Attach oil analysis results with method and unit, the baseline value, photos and lab report numbers.

6. Write the failure mode term, then the cause, then its certainty and the evidence for it.

7. Record the action and set a review date; on that date, record the outcome.

In an illustrative case, gearbox GB-12 at a paper mill shows moisture corrosion on an output bearing after 6,200 h on an ISO VG 320 mineral gear oil. Water reads 1,800 mg/kg against a new-oil value of 90 mg/kg (illustrative), and the site uses washdown hoses near the breather. The cause is recorded as water ingress through the breather, inferred, because no one has tested the breather itself (illustrative). The action is an oil change and a desiccant breather, and the review is set for 2,900 h later.

Why do lubricant failure records fail to help with the next similar case?

Lubricant failure records fail to help with the next similar case when they are written as stories, use different words for the same thing, leave out the operating conditions, or never record whether the fix held. Each of these breaks the comparison that the record exists for.

A story such as "the customer's gearbox failed after a hot summer, we suspected the oil" holds the facts a later engineer needs, but not in fields that can be matched. The same failure written by two engineers as "rusting" and "corrosion pitting" looks like two different cases to a search. A controlled vocabulary, such as the ISO 15243:2017 terms for bearings or the ISO 14224 failure modes for other equipment, removes that gap.

Missing conditions do the most damage. A past case without oil temperature, load or hours in service cannot be shown to be similar or different, so it either misleads or gets ignored. A cause written with no certainty reads as confirmed to the next reader, even when it was a guess on the day. A record that stops at the action hides failed fixes, and a failed fix is often the most useful thing a past case can teach.

The last cause is filling unknown fields with defaults: the design temperature, the nominal load or "standard" oil. Blanks are honest; defaults make unlike cases look alike.

Where this lubricant failure case record template does not fit

A lubricant failure case record template does not fit a root cause investigation report, a formal corrective action report for a customer, or a regulated incident record. It records one failure on fixed fields so that cases can be compared; it does not run the investigation or prove a cause.

A customer complaint that needs containment, root cause analysis and verified corrective action needs its own report in the format the customer or its quality system requires; attach the case record to it by case ID. An incident with injury, fire or a release to the environment follows the site's incident procedure and the rules of its jurisdiction, and this record holds only the lubrication facts for it.

The vocabulary sources have their own limits. ISO 14224:2016 was written for the petroleum, petrochemical and natural gas industries (ISO 14224:2016), and ISO 15243:2017 covers rolling bearings of standard bearing steels; its ISO life cycle stage is "International Standard to be revised" (ISO 15243:2017), so check the edition before relying on its terms. For gears, seals or hydraulic components, use the plant's own failure mode list and keep it fixed.

Lubinpla's similar case analysis compares the current case with past cases on axes it fixes before searching, and records kept on these fields can be added as work logs through bulk import, which accepts txt, md, csv, pdf and image files, so save the Word record as PDF or the register as CSV first.