Time and hours
Manufacturer schedules set time or operating-hour intervals. Inlet air, temperature, load and environment can change scope.
- planned and documented
- useful baseline
- not automatically condition-specific
Maintenance strategy
Improve plant availability:
maintenance is one element.
Preventive work can reduce fault risk. A credible availability strategy connects manufacturer schedule, condition data, process criticality, spares, service windows and reserve supply.
01 Principle
Availability does not come
from maintenance alone.
Regular maintenance can reveal deviations and reduce unplanned intervention. Availability also depends on design, inlet conditions, operation, measurement, spares, procedures, storage and backup.
No 24/7 promise: Critical supply requires the whole chain to be reviewed. A maintained single unit is not a redundant supply concept.
02 Three maintenance logics
Combine time, condition
and criticality.
Condition based
Measurement and diagnostics can reveal need if sensors, data quality, limits and interpretation exist.
- differential pressure and dew point
- purity, alarms and cycles
- trend rather than one value
Criticality based
The larger the failure consequence, the more important reserve, parts, safe restart and coordinated service windows become.
- assess process consequence
- define tolerable interruption
- test backup periodically
03 Unplanned downtime
Causes often sit
outside the main unit.
Without site data, frequency claims are not responsible. These categories structure recurring cases.
Supply
Power, compressed-air flow, pressure, temperature or quality outside limits.
Component
Filter loading, valve, sensor, drain, dryer, control or wear item.
Process
New users, peaks, changed purity, open bypass or unsuitable operation.
Organisation
Missing records, parts, ownership, alarm response or service window.
04 Improve availability
From alarm history
to an actionable plan.
This order prevents a focus on symptoms or one component only.
01
Current state
Equipment, interfaces, load, alarms, service and failure consequence.
02
Pattern
Compare time, load, temperature, pressure, operator action and last change.
03
Prioritise action
Assess safety, consequence, recurrence and effort.
04
Verify effect
Compare measurements, alarms or behaviour against the baseline.
05 Upgrade
Not every ageing system
needs full replacement.
An upgrade may mean maintenance, partial replacement, monitoring, treatment, storage, controls or complete replacement. Review the smallest action that reliably achieves the confirmed target.
Review an upgrade when …
- parts or manufacturer support end
- faults repeat for a structural reason
- demand or requirements grow materially
- energy, pressure or regeneration losses stand out
- required measurement or documentation is absent
Evidence first
- current output and actual load
- service and fault history
- parts availability and safety state
- target and measurable benefit
- outage and modification window
06 Common questions
Maintenance and upgrade
without buzzwords.
How can preventive maintenance reduce cost?
It can enable planned work, avoid consequential damage and reveal deviation before failure. The economic result depends on consequence, scope and actual condition.
What is predictive maintenance?
Condition-based or predictive maintenance uses suitable measurements, limits or models and a defined response process. The appropriate basis depends on the equipment and its criticality.
How can availability be improved?
Through suitable design, stable inputs, defined limits, maintenance, fast diagnosis, parts, trained response and a tested reserve route for critical processes.
When does an upgrade pay?
When measurable benefit in availability, capacity, quality, energy, parts support or operation justifies cost and outage. Compare it with maintenance and partial replacement.
Next step
Do faults repeat?
Start with the event timeline and exact equipment identification.
Alarm text, time, operating state, recent changes and maintenance history provide the basis for the next check.