Short answer: Roller replacement is a measurement task, not a visual one: act on two quantified criteria - rolling-surface diameter worn 6 to 8 percent of original, or flange wear-through deeper than 3 mm in one spot. A 0-300 mm caliper at two cross-sections in two directions per roller, walked in a fixed sequence, completes a full roller-set health check in about 20 minutes.

1. Why roller condition gets missed
Roller failure is gradual: 3 percent diameter loss is invisible, at 5 percent fuel use climbs and the machine starts pulling sideways, and by 8 percent the track-rail contact has degraded from surface to line contact - rail wear doubles, flange starts cutting the link sides. Fleets that watch track shoes alone discover late that the true cause is the roller; new shoes on worn rollers lose roughly 30 percent of their life from day one.
The second missed signal is temperature. After an hour of continuous travel, shoot each roller housing with an IR thermometer: any roller more than 15 degrees C hotter than its neighbors almost certainly has a bearing or seal problem. Temperature warns earlier than diameter - use both together.
2. Method and records
Tools: one 0-300 mm outside caliper and a diameter reference card listing new diameters and change-out limits per roller position. Measure at mid-face and near-flange, two directions each, record the minimum. Clean the measuring spot first - a mud shell inflates readings by 2-3 mm, the commonest source of false data.
Record on a top-view sketch: for each roller note rolling diameter, minimum flange thickness, housing temperature, seal condition. Two consecutive services side by side reveal wear rate - normal is 0.5-0.8 percent of diameter per 1000 hours; a roller that doubles its rate needs a duty investigation (the same hard crossing or bridge approach, usually).

3. Thresholds and pairing rules
Change-out criteria in priority order: diameter loss at or above 8 percent (6 percent on heavy/abrasive duty), flange wear-through at or above 3 mm, abnormal temperature with bearing wear confirmed on opening, or a seal that keeps leaking despite regreasing. Any one criterion justifies replacement. When a whole row of rollers sits above 6 percent at similar wear, replace the row together - a mixed set forces the new rollers to carry an overweight share and halves their life.
While replacing, check three companions: shaft runout with an indicator (over 0.3 mm, replace the shaft), the track-rail surface at the same positions, and the mounting-bolt torque. A bent shaft will reproduce the same one-sided wear pattern on a brand-new roller within 300 hours.
4. Make it a scheduled routine
Run the full roller set - measurement plus temperature - every 500 hours, merged with the oil-change visit to save a stop. Every 2000 hours, publish a trend sheet with wear-rate lines per roller. The trend sheet also drives spares: rollers run a 2-4 week procurement cycle, and a trend-based order lands the parts just in time instead of parking money in safety stock.
5. Roller replacement criteria
| Criterion | Standard duty | Heavy / abrasive | Action |
|---|---|---|---|
| Diameter loss | 8% or more | 6% or more | Replace, by row |
| Flange wear-through | 3 mm or more | 2 mm or more | Replace |
| Neighbor temperature gap | over 15 C | over 10 C | Open cavity, check bearing/seal |
| Shaft runout | over 0.3 mm | over 0.3 mm | Replace shaft |
Key takeaways
- Act on numbers: 8 percent diameter loss (6 on heavy duty), 3 mm flange wear-through, a 15-degree temperature gap, or 0.3 mm shaft runout.
- Temperature warns earlier than diameter - one hour of travel, then shoot every housing with an IR gun.
- Replace near-worn rows together; mixing new rollers into a worn row halves the new ones' life.
- Every 500 hours measure and shoot the full set; every 2000 hours publish the trend sheet and buy rollers against it.
Further reading
MG10000B 110-Ton Steel Track Undercarriage · Steel Track Undercarriage System 1-60 Tons · Maintenance knowledge column
Prepared by the MAGEON engineering team from custom project records and factory load-test data. Figures are engineering references; final selection should follow approved drawings and the actual duty spectrum.
English
简体中文