Undercarriage Wear Percentages: Why 80% Worn Is Not 20% of Life Left

An undercarriage report says a component is 80% worn. It is tempting to subtract that figure from 100 and conclude that 20% of its operating life remains. The subtraction is mathematically correct and mechanically incomplete. The reported percentage usually locates a measurement within an allowable dimensional range. Remaining service time asks how quickly the component will travel through the rest of that range under future conditions. One number cannot answer both questions unless an applicable wear relationship and a credible exposure forecast connect them.

An 80% reading can still help a planner identify a component approaching its documented limit. To place the next inspection, however, the direction and speed of the recent trend matter. Two components at the same percentage may need different attention because they arrived there along very different wear histories.

Wear position is not a clock

A dimensional wear percentage needs four identifiers before it can be interpreted: the component, the exact measurement point, the reference value for the unworn condition and the applicable service limit or wear curve. Remove any one of those and “80%” becomes ambiguous.

Consider a fictional feature that measured 100 units when new and has an applicable limit at 90 units. If its current measurement is 92 units, it has moved through eight of the ten units in the allowed dimensional range. On that defined basis, it is 80% through the allowance. The example says nothing about a real roller, link, bushing or sprocket, and the values must never be used as field limits. For this simple decreasing dimension, the calculation is:

(new dimension − current dimension) ÷ (new dimension − limit) × 100 = (100 − 92) ÷ (100 − 90) × 100 = 80%

This is a linear conversion of a dimensional allowance. A real inspection system may instead use a component-specific table or curve to convert its measurement. Read the report's definition before using this formula; the label “percent worn” alone does not establish how the number was obtained.

Even that calculation can be misleading if the references are wrong. A “new” value copied from a different configuration, a limit taken from another model, or a measurement made at a different datum changes the denominator. Wear may also affect more than one dimension, and the most advanced point can control the assessment even when an average looks acceptable. The applicable manufacturer procedure or validated inspection system must define the feature, tool, location and curve.

The remaining allowance has no time unit

“Allowable wear remaining” is another dimensional expression. In the fictional example, two units of the ten-unit allowance remain, so 20% of the dimensional allowance remains. That still is not 20% of time. It does not say that a component has 200 hours left because the first 80% took 800 hours. The earlier and later portions of a wear curve do not have to consume time at the same rate.

The rate between two inspections reflects the work performed during that interval as well as the component's wear behaviour. More travel, different ground or a changed mating component can make the next interval unlike the last. Measurement variation can also create an apparent change in rate. A reported percentage contains none of that history by itself.

The Komatsu undercarriage procedure manual, in its calculation of hours left, gives a physical reason for a nonlinear relationship. Its bushing outer-diameter example describes slower wear at a harder surface and faster wear as the wear progresses into a softer depth. Half the dimensional allowance can therefore take more than half the total wear life to consume. That example concerns the component and wear model in the manual; it does not establish a hardness profile for every aftermarket part. It does show why equal increments of lost material need not represent equal increments of time.

The component name also matters. A track roller does not share one generic wear curve with every roller position, machine and work environment. A percentage needs to stay attached to the particular roller position and the reference used to assess it. Averaging several positions can hide the one approaching its own limit.

A forecast needs a curve and exposure history

Two fictional measurement histories make the difference visible. Each row below is one equal interval of a consistently defined exposure measure. The values are invented percentages of dimensional allowance, not equipment test data or manufacturer wear curves.

Both end at 80%, but A changed by 10 percentage points in the latest interval while B changed by 35. Even their identical average since installation—20 percentage points per interval—conceals that difference. A straight average would give both the same projection precisely because it discards the sequence that distinguishes them.

There is a useful arithmetic thought experiment here. If each latest rate continued unchanged, A would take two more intervals to consume the remaining 20 percentage points. B would take about 0.57 of an interval, calculated as 20 ÷ 35. These are conditional results, not remaining-life estimates: nothing in the invented table establishes that either latest rate will persist. The contrast shows what the single “80%” reading leaves unanswered.

Before interpreting an actual history in this way, confirm that the measurement point, new reference, limit and conversion method stayed consistent. Pair every observation with its date and exposure reading. Note changes in duty, repairs, instrument or access that could explain a changed slope. Without that continuity, what looks like accelerating wear may be a change in how the measurement was taken.

With several comparable observations, a planner can see whether the measured rate has been stable, accelerating, slowing or too noisy to support a projection. A configured wear-management system may use model-specific curves and utilization assumptions to estimate a future point. TrackTreads describes forecasting based on configured wear curves together with utilization and operating conditions. That supports the need for a curve and exposure inputs; it does not guarantee that every fitted model predicts accurately or that the future will repeat the past.

Even a simple rate calculation needs a declared window. Dividing the latest dimensional change by the latest exposure can describe what happened between two inspections. Dividing total change since new by total hours describes an average over the entire history. The two rates can differ sharply on a curved wear path. A report should name which one it uses and show the underlying measurements rather than presenting the rate as a property permanently attached to the component.

Inspection intervals also affect what the curve appears to show. Widely spaced readings can hide a period of rapid change followed by slower wear. Very close readings may be dominated by measurement variation. There is no universal interval that solves both problems; the component, observed condition, applicable inspection program and consequence of reaching the limit determine how often another reliable point is worthwhile.

The next job can break the historical rate

Future duty is the second half of the forecast. Historical wear per engine hour may be a poor guide if the machine will move to a job with much more travel or different ground. An engine-hour meter can continue accumulating while the machine works largely stationary, whereas undercarriage exposure is strongly connected to travel and the conditions of that travel. If travel-specific data do not exist, the limitation should stay visible. Inventing a fixed travel ratio makes the projection look complete without adding evidence.

Measurement quality can overwhelm apparent trend. Suppose three readings differ by an amount similar to the expected change between inspections. A calculated rate will then be highly sensitive to small technique differences. Rechecking the datum, cleaning condition and repeatability may provide more value than fitting a more elaborate curve. More decimal places do not fix inconsistent measurements.

Component interactions can also make a single percentage a poor planning unit. A roller, idler, sprocket and track chain may occupy different positions in their allowable ranges, and the practical repair window may depend on how they work together. One component's dimensional percentage does not decide whether parts should be changed as a group. That judgment needs the applicable component limits, measured condition, mating wear patterns, repair scope and downtime plan.

Use the forecast to place the next observation

For planning, ranges or scenarios are often more useful than one countdown. One scenario can carry forward the recent measured rate and duty; another can show the implication of the recorded faster window. Both should expose the curve and exposure assumptions. The result can then place a reinspection before the projected limit instead of pretending to name the day the component will cease to be usable.

A remaining-hours estimate also needs to state what endpoint it predicts. Reaching a dimensional service limit is different from predicting functional failure, a safe operating period or an economical replacement date. Those decisions may depend on several components, machine availability, job requirements and applicable service guidance. A forecast should not be presented as permission to operate until the calculated hour.

In the two histories above, Component B's latest 35-percentage-point change deserves attention that the shared 80% figure conceals. A planner would need to understand that acceleration before carrying the latest rate forward—or dismissing it as noise. Component A presents a different recent trend. The percentage locates each component; the history explains why the next observation should not be scheduled from that percentage alone.

References

• TrackTreads undercarriage wear management
• Komatsu undercarriage procedure manual