Compressor alarms, trends and performance data

Alarms, trends and performance data: what each one is for

Three kinds of information come out of a compressor room, and they answer different questions.

Most plants have the first and little of the other two. That is why the same trip often returns: the reset clears the event, but nobody sees the pattern.

Designing alarms people will respond to

An alarm system that fires constantly trains people to ignore it. Good alarm design is about fewer, clearer alarms.

• Every alarm needs a consequence and a response. If nobody needs to do anything, it is information, not an alarm.
• Separate priorities. Trip (machine stopped), warning (approaching a limit), and maintenance due should look and route differently.
• Use specific text. “C2 high discharge temperature” is useful; “Compressor fault” is not.
• Set sensible limits and delays. A dew-point alarm that fires for every brief spike after a dryer restart becomes noise.
• Avoid floods. One power dip can trigger dozens of alarms; the first-out alarm is the one that matters.
• Review regularly. List the most frequent alarms each month and either fix the cause or fix the alarm.

Which points to trend

Record the points that explain compressor and system behaviour, not every register the controller offers.

• Header pressure after treatment, and delivery pressure at each compressor.
• Discharge temperature on each compressor (normally about 75–95 °C on an oil-injected screw).
• Run, loaded and fault state for each machine, with starts counted.
• VSD speed for variable-speed machines.
• Motor power or feeder kW.
• Header flow, where a meter is fitted.
• Pressure dew point after the dryer (typically about +3 °C for refrigerated dryers, −20 °C or −40 °C for desiccant).
• Compressor room ambient temperature, which matters a great deal in Indian summers.
• Differential pressure across separators and filters, where available.
• Maintenance events: oil change, filter change, repairs, with dates.

Choosing the right time window

Different problems happen at different speeds. Using one sample rate for everything either misses fast events or fills storage with data nobody uses.

A practical approach is fast sampling for short diagnostic campaigns and slower, long-retention logging for everyday history.

Correlate rather than guess: worked diagnostic patterns

When several trends sit on one time axis, causes become visible. Some patterns we see repeatedly:

• High temperature trips in the afternoon in April and May. Discharge temperature tracks room temperature. Look at cooler cleanliness, cooling air paths, hot-air recirculation and oil condition before raising any limit.
• Discharge temperature rising steadily over weeks at the same ambient. Fouled oil cooler, failing thermostatic valve or degraded oil.
• Header pressure falling while all compressors run loaded. Demand has exceeded capacity: new users, a large leak, or a machine delivering less than it should.
• Frequent starts and short loaded periods. Pressure band too narrow, too little storage or overlapping settings between machines.
• Flow during holidays or Sundays well above zero. Leakage. Compare it month by month.
• Power rising at the same flow. Specific power worsening: higher pressure, internal wear or more unloaded running.
• Dew point spikes at peak flow. Dryer overloaded or bypassed.

A simple procedure after any trip

1. Record the trip code, time, and which machine.
2. Pull trends for at least a few hours before the event, and for slow problems, several weeks.
3. Check what else changed: ambient, demand, other machines, power supply, recent maintenance.
4. Inspect the machine physically before resetting.
5. Record the finding and the action in the maintenance log.
6. Check the next few days of trends to confirm the fix worked.

For more on fault-finding see our page on breakdown diagnostics.

Performance indicators worth tracking

• Specific power, kW per m³/min of delivered air, where both power and flow are measured. Compare against the same machine over time, at the same pressure.
• Average header pressure, because each extra bar costs about 7% more power.
• Load factor, loaded hours divided by running hours, for each fixed-speed machine.
• Unloaded running hours, which consume power but deliver no air.
• Non-production flow, as a leak indicator.
• Trips per month and repeat trips by cause.

Indian operating conditions and what they mean for your data

Compressor data in Telangana, Andhra Pradesh and much of India has a strong seasonal signature, and trends should be read with that in mind.

• Summer heat. Compressor room temperatures in April and May can reach 40–45 °C or more, especially in rooms with poor ventilation or where hot cooling air recirculates. Discharge temperature, dryer dew point and power all move with it. Always trend ambient alongside machine temperatures, otherwise a seasonal rise looks like a fault and a real fault can hide inside the seasonal rise.
• Dust. Dusty sites around construction, cement, stone crushing and mining load inlet filters and coolers faster. Rising inlet filter or cooler differential, and a slow climb in discharge temperature, are often the first signs.
• Monsoon humidity. More water enters with the intake air. Dryer load rises, drains work harder and a failed drain shows quickly as dew-point excursions or water downstream.
• Power quality. Voltage dips and supply interruptions cause trips that are not compressor faults. Recording supply events, or at least noting them against trip times, prevents unnecessary repairs.

Comparing the same month across years, rather than one month with the next, is often the clearest way to see genuine deterioration.

Turning data into maintenance decisions

Trends earn their keep when they change the maintenance plan. A rising separator differential tells you to plan a separator change before oil carry-over starts. A cooler running steadily hotter at the same ambient tells you to schedule cleaning before the summer peak, not during it. Repeat trips on one machine point to an overhaul or airend inspection rather than another reset. Link these findings to your preventive maintenance plan so that data drives work orders.

Data governance: keeping the record trustworthy

A historian is only as good as its tags. Keep a tag list with names, units, scaling and source. Synchronise all clocks. Mark bad or missing data rather than interpolating it silently. Decide how long raw and summary data are kept and who can change tag definitions. After a controller replacement or firmware change, re-check the affected tags.

How Reddy Compressors helps

We are an independent multi-brand compressed-air specialist at Tarnaka, Secunderabad, serving industry since 2002, working on ELGi, Atlas Copco, Ingersoll Rand, KAESER, Chicago Pneumatic, Sullair, Doosan, Kirloskar, Gardner Denver and other makes. We are not an authorised dealer for any brand. We have tie-ups with all major and reputed brands. With more than 20 service engineers and technicians we cover all of Telangana and Andhra Pradesh, and we supply compressors, rentals and spares across India.

• We read controller event logs and trends during service and breakdown visits.
• We carry out temporary pressure, power and flow logging as part of air audits.
• We help set a practical alarm list and trend set for your station.
• We use the findings in service, overhaul and AMC work.
• We recommend and supply compressors, dryers and controls where the data shows a capacity or efficiency gap.

What to send us

• Make, model, serial and nameplate photos of each compressor.
• Photos or exports of the controller’s alarm and event history.
• Any trends or logged data, even screenshots.
• A description of the problem: when it happens, how often, what the operators see.
• Recent service records and any changes to the plant or air demand.

Frequently asked questions

Why does my screw compressor keep tripping on high temperature?

Common causes are a dirty or blocked oil cooler, poor ventilation or hot-air recirculation in the compressor room, low or degraded oil, a faulty thermostatic valve, a failed fan, or ambient temperatures above the machine’s design limit in summer. Trend discharge temperature against room temperature to see whether the problem follows ambient or is getting worse at the same ambient, then inspect the cooling circuit before resetting.

What data should I log on an air compressor?

Log header and delivery pressure, discharge temperature, run, loaded and fault state, starts, VSD speed where fitted, motor power, header flow if metered, dryer dew point and compressor room temperature. Add maintenance events with dates. Sample fast for short diagnostic periods and slower for long-term history, and make sure all devices share one synchronised clock.

How can trends help find compressed air leaks?

Leaks show as air flow, or compressor loading, when production is stopped. Trend flow or loaded time during Sundays, holidays and break periods and compare it month by month. A rising baseline means leaks are growing. Trends cannot locate leaks, so follow up with an ultrasonic leak survey and a repair register, then check the trend again after repairs.

What is a good alarm setup for a compressor room?

Keep it short and meaningful. Separate trips, warnings and maintenance reminders; use clear text naming the machine and cause; set delays to avoid nuisance alarms; and give each alarm an owner and a first response. Review the most frequent alarms every month and fix either the cause or the alarm setting.

What does specific power tell me about my compressor?

Specific power is the electrical input per unit of delivered air, usually in kW per m³/min. Tracked over time at the same pressure, a rising value means the machine or station is becoming less efficient, through wear, fouling, more unloaded running or higher pressure. It needs both a reliable power meter and a flow meter, and comparisons are only fair at the same pressure and conditions.

How long should compressor data be stored?

Keep detailed short-interval data for weeks to a few months for diagnostics, and hourly or daily summaries for at least a year so you can compare seasons. Summer and monsoon conditions in India change compressor behaviour, so year-on-year comparison is valuable. Decide the retention and who owns the archive when the system is set up.

Related pages

• Compressed-air controls and monitoring hub
• Condition monitoring
• Remote compressor monitoring
• Power and flow logging
• Specific-power assessment