A packaged system can look finished and still arrive with the wrong control sequence, an inaccessible strainer, or no documented proof that it meets the purchase requirements. Finding those problems after the rigging crew leaves puts the correction work in the plant’s schedule.
An industrial equipment skid factory acceptance test should demonstrate that the package matches the approved design, operates within agreed test conditions, responds correctly to faults, and has the records needed for installation and startup. Anything the supplier cannot demonstrate should be identified before testing—not explained away afterward.
The useful question isn’t simply, “Did it run?” It’s, “What did this test prove, and what still needs verification at our facility?”
Define the FAT Before the Purchase Order Is Released
A factory acceptance test, or FAT, is a documented inspection and test performed before equipment leaves the supplier’s facility. Its scope depends on the purchase agreement. A brief motor run and a witnessed package performance test are not equivalent, even if both proposals call them a FAT.
For packaged pump systems, blower packages, vacuum systems, and other process skids, put these items in the purchase scope:
Design basis: Required duty points, operating range, fluid or gas properties, utilities, environmental conditions, and equipment boundaries.
Approved references: Equipment lists, drawings, piping and instrumentation diagrams, electrical schematics, and control descriptions, with revision numbers.
Test methods: Test medium, available load, operating duration, measurement locations, and permitted simulations.
Acceptance criteria: Numerical limits or observable pass/fail requirements for each test.
Responsibilities: Who prepares the procedure, witnesses testing, resolves discrepancies, and authorizes shipment.
Commercial arrangements: Witness notice, retesting costs, documentation deadlines, and treatment of unfinished work.
Approve the procedure before the witness visit. That visit should be spent observing tests, not negotiating what “acceptable vibration” means.
Verify the Assembled Skid Against the Approved Design
Start with what was actually built. Record equipment tags, model information, serial numbers, materials documentation where specified, motor ratings, instrument ranges, valve arrangements, and connection sizes.
Compare the physical assembly with the approved drawings. Confirm flow direction, check-valve orientation, vents, drains, isolation points, instrument locations, and relief provisions where required. Record substitutions and obtain approval rather than treating them as paperwork cleanup.
Look at maintenance access while everything is visible. Can someone remove a strainer basket, change a mechanical seal, or service a motor without dismantling unrelated piping? Are electrical enclosure doors obstructed? Can operators read gauges from the intended access side?
Check dimensions, connection locations, shipping weight information, lifting provisions, and foundation interfaces against the installation plan. A skid that passes its operating test can still create expensive field changes if its connections don’t match the plant layout.
Demonstrate Mechanical Integrity and Operating Performance
Separate pressure testing from running tests
Pressure-boundary testing and operating leak checks answer different questions. Define the applicable test requirements, medium, pressure, duration, boundaries, and acceptance limits using the governing design requirements and manufacturer instructions. Not every component can tolerate the same test pressure.
Pressure testing belongs under an approved procedure performed by qualified personnel. Pneumatic testing carries substantial stored-energy hazards and should not be substituted casually for hydrostatic testing.
During the operating run, document leaks, bearing temperatures, motor current, vibration where specified, lubrication condition, and abnormal noise. Include seal-support, cooling, and lubrication systems that are part of the package.
Measure performance at meaningful conditions
For a pump skid, useful evidence may include flow, suction and discharge pressures, speed, fluid temperature, and input power where required. Pressure readings need enough context to establish pump differential head; a discharge gauge alone doesn’t prove pump performance.
For blower, compressed air, or vacuum packages, define inlet conditions, pressure or vacuum measurement basis, gas temperature, flow reference conditions, and power measurements. “Flow” without stating actual or standardized conditions can lead to conflicting interpretations.
Test the contracted duty points and relevant operating range—not just whichever condition the shop loop can reach. Agree on stabilization time and run duration. A short demonstration may confirm rotation and basic operation without establishing sustained thermal behavior.
Record instrument identification, calibration status, measurement accuracy, and agreed tolerances. If measurement uncertainty overlaps an acceptance limit, the procedure should explain how that result will be evaluated.
Prove the Controls, Not Just the Motors
A skid may run correctly in manual mode while its automatic sequence remains unfinished. Test the approved operating sequence from startup through normal shutdown, including transitions between operating modes.
Permissives: Conditions that must be satisfied before equipment can start.
Automatic operation: Pressure, flow, level, or temperature control under available test loads.
Multiple-unit operation: Lead/lag changes, standby starts, and response to an unavailable machine.
Fault response: Specified trips, alarm messages, sensor failures, and communication loss.
Power interruption: The agreed shutdown, recovery, and restart behavior.
Plant interfaces: Remote commands, status signals, engineering units, scaling, and network data mapping.
For variable frequency drives, verify approved operating limits, acceleration and deceleration settings, and control ownership. The plant and skid controller should not both be trying to regulate the same variable without a defined strategy.
Use approved simulation methods where creating the actual fault would be unsafe or damaging. Record whether a test exercised the real field device, the wiring and controller, or only software logic. Those are different levels of verification.
Safety-related functions need validation appropriate to their design requirements. An ordinary sequence demonstration does not replace that work. Testing must follow facility safety procedures, manufacturer instructions, and applicable regulations.
State What the Factory Test Cannot Prove
A FAT usually runs with shop utilities and a temporary test circuit. It does not automatically prove performance with plant piping, actual process fluid, or the final control network.
A water test can demonstrate mechanical operation of a chemical pump package, but it may not establish performance with a viscous product. Differences in viscosity, density, vapor pressure, solids, or temperature may require approved corrections, separate testing, or site verification.
Consider a hypothetical North Mississippi plant buying a transfer-pump skid. The supplier’s flooded shop suction allows a successful water test. At the plant, a long suction line and restrictive strainer reduce available suction head. Cavitation during startup would not, by itself, mean the factory test was false; it could mean the test never represented the installed system.
Likewise, a compressed air dryer tested under mild shop conditions has not necessarily demonstrated performance in a hot, humid Mid-South mechanical room. Document the expected ambient and inlet conditions and review the applicable manufacturer ratings.
Assign every unproven requirement to a named follow-up activity, usually a site acceptance test, or SAT. Site testing should address installed piping, utilities, field wiring, communications, and actual process interaction.
Require a Usable Test Record and Shipment Decision
A signed cover sheet saying “FAT complete” is not enough. The final package should include:
The approved procedure with actual readings, results, dates, and witness identification.
Test conditions, instrument records, trend data where required, and clearly identified simulations.
Updated drawings, equipment records, operating manuals, and maintenance requirements.
Approved controller and operator-interface backups, drive settings, and access arrangements.
A discrepancy list showing corrective action, responsible party, retest requirements, and closure evidence.
Shipping preservation instructions, including draining, drying, connection protection, and storage requirements.
Separate shipment-blocking defects from items the buyer explicitly accepts for later completion. Unresolved safety issues, failed contractual performance, and missing required protective functions should not disappear into a general punch list.
If a correction changes wiring, software, or equipment behavior, repeat the affected tests. Define shipment release separately from final contractual acceptance so neither party has to guess what a signature means.
Bottom Line
A useful industrial equipment skid factory acceptance test connects each purchase requirement to a demonstration, a measured result, or a documented site-verification task. Define those connections before ordering. Then release the skid based on evidence and agreed exceptions, not appearance or schedule pressure.
For a packaged-system inquiry in Tennessee, Arkansas, or Mississippi, send Process & Power your duty requirements, process conditions, utility information, drawings, controls expectations, and proposed FAT scope. That gives the equipment discussion a practical starting point.
Call Process & Power, Inc. at 901-362-5500 or visit us at 1721 Corporate Avenue, Memphis, TN 38132 for help with industrial air compressors, pumps, blowers, vacuum systems, and compressed air equipment and service throughout Memphis and the surrounding area.
