Reliability and maintenanceRM-003

Continuous gearbox monitoring at Radotín

Ceskomoravsky Cement · Radotin cement plant, Czech Republic · reported 2021

Use this record when

The decision this case can inform

Use this record when temperature or oil-flow alarms detect gearbox damage too late and the plant is selecting measurements tied to the actual failure signature.

Evidence scope

Named SKF supplier case with a qualitative operational outcome. No before/after downtime, alarm-performance, or economic series is published.

Source-supported facts

What the public record actually establishes

4 sourced points
  1. F1

    Two grinding-mill gearboxes failed within a few months, each stopping the line for several days and reducing plant capacity by half.

  2. F2

    Existing temperature and oil-flow sensors indicated problems only after damage was advanced.

  3. F3

    Permanent vibration sensors detected bearing damage through frequency-specific signatures.

  4. F4

    The plant and supplier report that earlier warning enabled planned repair and led to a long-term remote diagnostic support arrangement.

Structured interpretation

Facts and reported results are kept separate from the lesson a plant may choose to test.

01

Operating context

Recurrent grinding-drive gearbox failures stopped a line for days and cut capacity in half; temperature and oil-flow detection was late.

02

Intervention or finding

Permanent vibration sensors and remote diagnostics were installed.

03

Documented result

Plant and supplier report earlier detection, less damage exposure, plannable repairs, and a long-term support contract; no quantified availability delta.

04

Plant interpretation

Monitor the variable that reveals the failure mode and connect alarms to prepared maintenance decisions.

05

Transfer boundary

Supplier-reported and qualitative; no before/after downtime series.

Before applying the lesson

Questions to verify at your plant

These are decision checks, not operating instructions. Resolve them through local risk assessment, technical review, and authorization.

  1. 01

    Which measurable signal appears early enough for the target failure mode and operating speed?

  2. 02

    How will load, speed, process state, and sensor health be captured so analysts can distinguish a fault from operating variation?

  3. 03

    What lead time is required to plan parts, labor, access, and a controlled outage?

  4. 04

    Will alarm precision, warning time, avoided damage, and false positives be recorded after deployment?