Short answer: Most compressor failures trace back to the first week, not the tenth year. A structured 72-hour commissioning check — rotation direction, pressure band, condensate behaviour, and a recorded baseline of power and flow — catches nearly all of them, and gives you the reference numbers that make every future efficiency claim verifiable. Record at least three numbers in the first 72 hours: input kW at known load, FAD at working pressure verified per ISO 1217:2009 acceptance tolerances, and specific power (kW ÷ m³/min).
By Johnny Wayne, General Manager, AirSpace Machinery Co., Ltd. — 20 years in industrial compressed air systems. Reviewed by our engineering team.
A new compressor arrives, it starts, air comes out of the line — and everyone moves on. That is how plants end up with machines running for years with a wrong rotation direction on the standby unit, a pressure band nobody remembers setting, and no record of what “healthy” ever looked like.
The first 72 hours are cheap. Everything after them is expensive. Here is the checklist we use, in the order it should happen.
Day 0 — what should you check before the first start?
These checks take an hour and prevent the failures that take weeks to diagnose:
- Room, not just machine. Clearance around the package for airflow and service access; ventilation sized so the heat the machine rejects can leave the room. Every kilowatt of electrical power going in comes out as hot air, so a 75 kW machine heats a small room like a furnace.
- Oil level and correct oil grade. The wrong oil is the fastest warranty-voiding mistake in this industry. Check what the manual specifies, not what the previous machine used.
- Electrical tightness and rotation direction. Terminals torqued, earth connected, and — on first power-up — a rotation check before running loaded. A screw running backwards is damaged within minutes; on two-machine rooms, verify both, because supply phases do not always match the drawing.
- Downstream ready. Receiver isolation open, dryer and filters rated for the machine’s flow (not inherited from the old unit), condensate drains connected and functioning. If the air touches food, pharmaceutical product or a laser lens, confirm the ISO 8573-1:2010 class you actually need against the machine’s rated purity.
If you are installing an all-in-one skid unit — compressor, dryer, filters and receiver on one frame — most downstream items are pre-engineered, and the checklist collapses to the room, the power and the drains.
Hours 1–2 — what should you watch on the first run?
Start the machine unloaded, then watch it earn your trust:
- Rotation direction matches the arrow — again, because this is the one that eats machines
- Oil pressure builds and stabilises; discharge temperature climbs smoothly to its normal band, with no oscillation
- No rattles, cavitation noise or metallic ticking
- The controller’s pressure band matches your plant’s real requirement. Machines ship with a sensible default, but a default is not your plant. A band that is too wide makes a fixed-speed machine cycle; too narrow and a VSD machine hunts. Setting this one parameter correctly is worth more than any accessory you can buy.
- Safety valve proves, emergency stop actually stops, and the machine unloads cleanly at the top of the band
Hours 2–24 — what should you log?
Run it at working pressure and log at least five readings — hourly for the first 3 hours, then every 4 hours. Typically the first 24 hours tell you whether the problem is the machine or the installation:
- Input power (kW) at a known load
- Discharge and oil temperature — stable, not drifting
- Pressure drop across new filters — a fresh filter should read low; if it does not, the sizing is wrong, not the filter
- Condensate volume. Large volumes in the first day are normal in humid climates and tell you the drains are earning their keep; a dry trap in humid weather means a blocked drain, and water is heading downstream
- A leak walk of the new pipework. New installations leak at every joint they have not pressed; soap solution or an ultrasonic gun finds in an hour what wastes money for years
Day 2 — which baseline numbers should you record?
This is the step almost everyone skips, and the one with money attached. Write these down, dated:
- FAD at working pressure, verified against the nameplate. Delivery is tested per ISO 1217:2009 acceptance tolerances, so know what that tolerance is before you treat a difference as a fault.
- Specific power: input kW divided by delivered flow. This single number is your machine’s efficiency fingerprint.
- Noise level at working distance, if your environment cares.
Why it matters: every future claim — energy savings, degradation, the case for a VSD upgrade — is measured against this baseline. When a supplier tells you a machine saves 35% against fixed speed, that figure is a planning benchmark rather than a promise; your own baseline is what turns it into your number. It is also exactly the input our free kWh forecast and system review uses.
Day 3 — what does a proper handover include?
Commissioning ends with people, not pressure readings:
- Operator trained on the controller: start/stop, band adjustment, alarm history — not just the green button
- Service schedule in the calendar (oil, filters, separator) with the grade and part numbers written where the machine lives
- Minimum spares on the shelf: oil, filters, drains — sized to your distance from the service depot
- Logbook started, with the Day-2 baseline inside, dated and signed
Two machines installed in the same week, from the same factory, will be in radically different condition after five years — and the difference is almost always this checklist.
What does a missing baseline cost you later?
Our published project data shows what the absence of a baseline looks like from the other side. In one 75 kW PMV VSD retrofit case study, the plant arrived at the upgrade conversation with pressure swinging ±0.5 bar, compressed air costing over 3,200 USD a month, and maintenance averaging 850 USD a month — numbers that had to be reconstructed from bills rather than read off a logbook. A second documented project, a 34% energy reduction with a 14-month payback, turned on having measured input power before and after rather than trusting a supplier’s estimate.
In practice the pattern repeats on site: plants with a signed Day-2 logbook can prove a saving in a week, and plants without one spend months arguing about whether anything changed. That is the whole argument for spending an hour with a clamp meter in week one.
Key takeaways
- Record three numbers in the first 72 hours: input kW at known load, FAD at working pressure, and specific power (kW ÷ m³/min).
- Verify rotation direction before running loaded. A reversed screw is damaged within minutes, and supply phases do not always match the drawing.
- Set the pressure band on day one. A default band is not your plant’s band, and it decides how the machine cycles for the next decade.
- A dry condensate trap in humid weather is a fault, not a saving — water is heading downstream.
- Every future saving claim is measured against the Day-2 baseline. Without it you cannot prove a 35% benchmark either way.
Frequently asked questions
How soon can a new screw compressor run at full load?
After a rotation check and a clean unloaded start, most modern packages take full load within the first hour. The 72 hours are for verification and baselining, not break-in — modern screw packages do not need an extended break-in period.
What is the single most important number to record in week one?
Specific power — input kW divided by delivered flow (FAD). It is the machine’s efficiency fingerprint, and every future saving or degradation claim is measured against it.
Who should do the commissioning?
An authorised service technician, on site where possible. For export markets where travel is impractical, structured remote commissioning with a video walkthrough and a supervised checklist works — but someone at the site must own the rotation-direction and electrical checks in person.
What voids a compressor warranty fastest?
In order: wrong oil grade, reversed rotation, and starved ventilation. All three are Day-0 checks that take minutes.
The compressor room gets very hot in the first days — is that a problem?
Heat is normal; accumulating heat is not. Every kW in comes out as heat, so if the room keeps climbing, ventilation is undersized and the machine will start derating or tripping on high temperature. Fix the airflow before summer finds it for you.
Do I need to test air purity at commissioning?
If the air touches product, yes. Commissioning is the cheapest moment to sample against ISO 8573-1:2010 and confirm the class you specified is the class you got — oil carryover and dew point problems are far cheaper to catch in week one than in a product recall.
Sources and standards
- ISO 1217:2009 — Displacement compressors — Acceptance tests. The basis for verifying FAD against a nameplate figure, including the acceptance tolerance.
- ISO 8573-1:2010 — Compressed air — Contaminants and purity classes. Relevant wherever air touches product, packaging or a laser lens.
- ISO 11011:2013 — Compressed air — Energy efficiency — Assessment. The method behind a proper measurement baseline.
- EU Machinery Directive 2006/42/EC — the directive under which CE marking is applied to machinery placed on the EU market.
- Our published case studies — 75 kW PMV VSD installation, Vietnam and 34% savings against a European-tech brand.
About the author
Johnny Wayne (Wei Zhuang) is Managing Director of AirSpace Machinery Co., Ltd., the Shanghai manufacturer behind ChinaCompressor.org. The company builds oil-injected and Class 0 oil-free screw compressors from 2 to 540 HP (1.5–400 kW) on a 4,000 m² production facility with 20 years of engineering history behind it, using BAOSI 5:6 asymmetric rotor technology. Products carry CE marking and ISO 9001:2015 quality certification, and we sell through distributors rather than around them. This checklist is the one our own engineering team works through on commissioning visits.
Need a duty point checked, or a commissioning walkthrough? Send pressure, flow, hours and your electricity tariff to sales@chinacompressor.org.
📅 Last updated:
⚡ Is Your Factory Bleeding Cash?
Most fixed-speed compressors waste $3,600/year in "Unload Tax." Test your waste level in 30 seconds.
RUN THE ROI TEST →Ready to See How Much Your Facility Can Save?
Fill out the form below for a Free Energy Assessment.
