When a critical conveyor stops, the fastest path back to production is usually a spare part. Yet many warehouse operators discover only after the pallet is lifted, the guard is refitted and the fault re-occurs that the newly “commissioned” spare was never truly accepted. Commissioning and acceptance are not simply the paperwork that accompanies a delivery. They are the technical checks that confirm a spare part is the correct item, in genuine serviceable condition, and capable of performing its function in your specific application. This article provides a neutral, engineering-led checklist for commissioning and acceptance of critical spare parts in warehouse automation systems, with a focus on observable evidence, diagnostics and decision boundaries.
Why Acceptance Is Separate from Delivery #
In a typical warehouse, incoming goods are checked against a delivery note, then moved to a store. That process verifies quantity, not suitability. A critical spare part can be correctly named on the delivery note yet be the wrong variant, the wrong revision, or damaged in a way that is not visible in packaging. Acceptance for critical spares must verify three separate attributes:
- Identity – the part matches the OEM specification and the site’s bill of materials for the installed equipment.
- Condition – the part is free from damage that would shorten its service life or cause secondary failures.
- Function – the part performs as designed when subjected to static and, where possible, dynamic checks.
Acceptance only ends when all three are confirmed and the result is recorded. This defines the boundary between a stock item and a commissioned spare. Site procedures, OEM documentation, lock-out/tag-out rules and the judgement of a competent engineer must always take priority over any generic checklist.
Defining Critical in a Warehouse Automation Context #
Not all spares deserve the same commissioning depth. A roller used in hundreds of positions is often standard stock, replaced quickly, and has low risk of causing collateral damage. A critical spare, by contrast, has three common characteristics:
- It is a single point of failure for a production-critical asset – for example, a sorter drive motor or a PLC safety input module.
- Its lead time is long relative to acceptable downtime, meaning you cannot afford to discover a problem after installation.
- Its failure mode can damage other components – for instance, a mis-selected drum motor that overloads the inverter, or a gearbox with incorrect oil fill that seizes and twists its coupling.
In practice, critical spares are usually found in the material flow path, in the electrical control circuit, or in the safety chain. A component that is part of the safety chain is not merely a spare part; it is a piece of safety-related equipment and must be handled with extra care. Final decisions on criticality should be made by the maintenance and engineering team together with the controls team, and the list should be reviewed whenever a major upgrade is installed.
The Pre-Commissioning Documentation Review #
Before any physical inspection is performed, documentation must be reconciled. A common error is to check only the part number printed on the device. The printed label can be misleading, especially when an OEM has superseded an old part number with a new one.
Collect and compare the following whenever available:
- Purchase order line and the delivery note line.
- OEM part number, its supersession history and any applicable upgrade bulletins from your own maintenance records.
- Manufacturer date code, batch code and serial number (where present).
- Firmware revision or software version for drives, controllers and smart sensors.
- Nameplate data such as voltage, current, speed, torque, sensing range or output type.
- Any certificates or OEM test documentation supplied with the part.
When an OEM part number has been superseded, the supplier should state the interchangeability basis. It is not enough that the mounting holes line up. The new variant must have equivalent electrical characteristics, thermal limits and control interface behaviour. Where firmware is not the latest release, this should be noted before store entry, not discovered after commissioning causes a comms fault on the network.
Inspection and Dimensional Verification #
Once documentation is aligned, the physical part must be inspected. Do not simply observe packaging. Open the part, examine the mating surfaces, and compare key dimensions to the OEM drawing or to a known good part removed from a line that is being overhauled. The goal is to identify split threads, bent shafts, deformed housings, damaged keyseats and incorrect shaft lengths, all of which cause vibration, misalignment or binding long before the part reaches its normal wear limit.
The table below summarises typical checks for common warehouse spare part families. The exact dimensions and tolerances come from the OEM drawing, not from this table.
| Part family | Key checks | Tool typically used | Common acceptance discrepancy |
|---|---|---|---|
| Motor and gearbox | Shaft runout, keyseat width, flange pilot diameter, bolt hole spacing, oil fill level | Dial indicator, vernier caliper | Undersized keyway, damaged shaft end thread, wrong flange size for the mounting adapter |
| Conveyor drum and rollers | Roller length, axle length, bearing housing diameter, hex shaft orientation | Steel rule, caliper | Roller is 8 mm too short after thermal expansion is considered; |
| Drives and inverters | Rated current, heat sink condition, terminal labels, control plug pin layout | Visual inspection, ohmmeter on connectors | Same frame size but lower output current rating than required |
| Industrial sensors | Thread size, sensing face material, cable length, connector keying, output logic | Thread gauge, caliper, local wiring diagram | PNP sensor supplied where NPN is required |
| Couplings and pulleys | Bore diameter, keyseat orientation, bushing range, belt section compatibility | Caliper, bore gauge | Pulley pitch diameter correct but hub bore is metric when the shaft is imperial |
| Control modules | Module type, terminal layout, backplane connector condition, firmware label | Visual inspection, ohmmeter checks against site I/O list | Module is the correct family but is the obsolete revision without thermal derating |
Dimensional verification is not over-engineering. A sorter wheel that is half a millimetre larger than specification may feel acceptable by hand but will create pre-load that overheats the bearing in the first shift. The cost of the measurement is minutes; the cost of a re-installation is hours plus potential damage to the counter-rotating rollers.
Electrical and Electronic Verification #
Electrical spares need a different inspection routine because damage is often invisible. Begin with the evidence you can gather without applying power.
- Check all connectors for bent pins, pushed-back terminals, contamination or moisture ingress.
- Inspect heatsinks and fan grilles for blocked airflow paths and trapped debris.
- Measure continuity on power and earth paths where the OEM permits, using the appropriate test equipment and observing electrical isolation rules.
- For motors, where site procedure and OEM documentation allow, measure winding resistance between phases and each phase to earth. Record the actual readings, not simply “pass”.
- For sensors, compare the connector pinout and output logic with the site’s wiring diagram.
Electronic components must be handled using anti-static precautions. A drive board can be damaged by static before it is ever connected. Use a grounded workbench, dissipative mat and wrist strap when inspecting modules outside their packaging. If the part is stored in an anti-static bag, do not remove it unless it is ready for inspection or installation.
Do not assume that a newly supplied part is electrically sound. Poor handling in transit or exposure to condensation in a warehouse doorway can cause intermittent faults that only appear after installation. The acceptance step should record resistance results and the date of the measurement. If the part is placed into store for two years, those records become the baseline for a later re-test.
Functional Testing and Test Rigs #
For many mechanical spares, dimensional checks and a slow manual rotation are enough to confirm function. For motors, drives, sensors and control modules, further testing is advisable but must be performed in a controlled manner, never by altering safety circuits or by bypassing interlocks. A dedicated bench test is the safest approach.
A simple test rig for a conveyor motor might consist of a controlled power source, a local start/stop station, and a means of observing rotation direction and current draw. For a sensor, the test rig may be a DC supply and a small metal or reflective target. For a drive, the test rig may be a spare motor that is connected only under lock-out conditions. In all cases, the following principles apply:
- Follow the OEM’s commissioning instructions step by step.
- Use the site’s lock-out/tag-out procedure and verify the test rig is isolated before changing connections.
- Record the test parameters: supply voltage, observed current, speed, output state, response time.
- Define a pass/fail boundary in advance, not after the test result is known.
An example of a functional test boundary for a photo-eye sensor might be:
- Power indicator lights at rated voltage.
- Output changes state when the target is placed at the specified sensing distance.
- Output returns to its original state when the target is removed.
- No false triggering when a similar object is placed outside the sensing range.
When a part passes this level of testing, you have far stronger evidence that it will work in the machine. When it fails, you have saved yourself an unplanned breakdown later.
Evidence Collection and Record Keeping #
Acceptance without documented evidence is an opinion. The commissioning record should be retained alongside the spare part’s identification in the CMMS or maintenance file. Include the following:
- Photographs of the part, including the serial number and any unique identifying marks.
- Photographs of any dimensional measurements, with the tool visibly readable in the image where possible.
- Electrical readings and their test conditions.
- Functional test results, including the pass/fail boundary and who performed the test.
- The store location and any condition flags, such as “firmware to be upgraded before installation”.
This record is not bureaucracy. It becomes the evidence base for later root cause analysis. If a part fails after installation, you can quickly establish whether it was a latent acceptance defect, a storage-induced degradation, or an installation error. Without the record, you are forced to reproduce the entire verification process while the line is down.
Common Interpretation Errors #
Experienced engineers make mistakes, and the mistakes tend to fall into repeated patterns. Awareness of these patterns reduces the chance of acceptance errors.
- Date code confusion. Manufacturers use different date code formats. A code that starts with “D” may mean the fourth decade, not December. If a part has been on the shelf for years, electrolytic capacitors may be degraded even if the part is new in its box.
- Cosmetic similarity. Two drives may share a case and a front panel but differ in output current and control firmware. Never rely on appearance alone. The nameplate is the reference, and the nameplate must be cross-checked against the OEM data.
- Sticker-over-sticker. A relabelled part can hide a supplier’s rework or a firmware update. If the underlying label is not visible and verifiable, treat the part as requiring full functional testing before installation.
- Power-on means pass. A lamp illuminating proves only that the lamp is getting power. The output logic, the sensing repeatability and the response time must be confirmed.
- Confusing remanufactured with new. There is nothing wrong with a genuine refurbished part, but it must be labelled as such and accepted to the same functional standard, not assumed to be new. The opposite error is also common: rejecting a perfectly good refurbished part because it lacks the original gloss.
- Ignoring the system context. A full-load motor test in isolation cannot replicate the inertia of a long conveyor. The acceptance test should be noted as a short-duration verification, not a full thermal validation.
Decision Boundaries and Escalation #
Acceptance results in one of three decisions. These should be recorded explicitly rather than implied.
- Accepted. The part meets all identity, condition and functional criteria and can be placed into controlled stores or installed in the asset.
- Accepted with conditions. The part is serviceable but requires a defined action before installation. Examples include a firmware upgrade, replacement of a damaged connector hood, or cleaning of a heatsink. The condition must be written on the part and in the CMMS, otherwise it will be forgotten.
- Rejected. The part is not suitable and must be quarantined, returned or disposed of in line with site procedures.
Rejection triggers include any visible crack in a housing, damaged mating threads, incorrect voltage/current rating, missing firmware, evidence of moisture ingress in electronics, or a dimensional mismatch. Do not attempt to fit a rejected part simply because the line is down and there is no alternative. A part that has already shown an acceptance fault will likely fail during the next restart, and it may take the OEM’s only replacement with it.
Escalation is appropriate whenever there is conflict between the delivery documents and the physical evidence, or when the site’s engineering team cannot agree on the disposition of a part. Escalate to the maintenance manager or reliability lead before the part is placed into bin storage. Document the decision and its rationale so that future commissioning decisions are consistent.
Maintenance and Reliability Implications #
The commissioning and acceptance of critical spares has a direct effect on equipment uptime and on the quality of maintenance data. Consider the classic repeat-fault scenario: a conveyor drive fails, is replaced, and fails again after a few hours. The maintenance team concludes the machine has a structural issue and begins an expensive investigation. In fact, the first fault was legitimate, and the second fault was caused by a spare motor that had been accepted on the basis of a delivery note but had suffered a damaged keyseat in transit. The evidence existed – dimensional measurement would have found it – but it was not collected at acceptance. The result is wasted labour, a false diagnosis and a loss of trust in the reliability process.
A disciplined acceptance loop also helps the failure coding system. When a failed spare is returned to the supplier for warranty, the acceptance record proves how the part was handled before installation. This protects the site against
