From Factory Witness Tests to Site Energization
Why the report that leaves the plant still matters on the day you close the breaker
When medium- and high-voltage power transformers ship with a certified factory test report, that packet is not paperwork for the file cabinet. It is the first half of the commissioning story. The second half happens on site—receiving, installation, field testing, and energization. Owners who treat those as separate worlds pay for the gap twice: once in schedule risk, and again when a factory Accepted line item does not match what the field can prove.
Switchgear lineups, UPS systems, and other skidded gear also leave plants with factory acceptance tests (FATs). Those packages follow different standards and test menus than liquid-filled transformers. This post focuses on the transformer witness / certified test package—the IEEE C57.12.90-class suite most owners see on power transformers—and how independent commissioning (Cx) should carry that serial-number baseline all the way to a controlled energization. The same handoff logic applies to other equipment classes; only the factory chapter titles change.
At Quinn Electrical and Instrumentation (QEI), we work both sides of that handoff: reading what the manufacturer actually proved, then making sure the project can prove the system before anyone closes a breaker.
Factory witness is not a rubber stamp
A witness test (or certified factory test package) documents that the manufacturer ran defined electrical tests under a controlling standard—commonly IEEE C57.12.90 for liquid-filled transformers—with measured values compared to design or guarantee bands. A typical transformer factory package includes:
Ratio — turns ratio matches nameplate and taps; field Cx still cares because wrong ratio or tap wiring shows up as voltage problems under load.
Winding resistance — conductors and connections are correct; loose connections become hot spots and I²R surprises.
Polarity, angular displacement, and phase sequence — terminal relationships are right for paralleling; mistakes become safety and outage events.
Insulation resistance — timed DC health check; shipping and install create a different moisture/damage risk than the plant saw.
No-load losses and excitation — core losses and magnetizing current at rated voltage (often also 110%); baseline for efficiency and later troubleshooting.
Load losses and impedance — copper losses and %Z at a reference temperature and tap; %Z drives fault current and voltage drop.
Insulation power factor (and capacitance) — moisture / dielectric condition; a dry plant unit can still pick up moisture in storage.
Applied-voltage and induced-voltage tests — dielectric withstand between windings/ground and turn-to-turn integrity.
Lightning impulse (when required) — BIL withstand for the insulation class; not every job runs impulse.
A clean Accepted on no-load losses, load losses, total losses, and impedance is necessary—but it is not sufficient for energization. It means this serial number left the factory within the bands the report claims. It does not automatically mean the unit was received undamaged, stored dry, landed on the right pad, terminated to the right phases, or integrated into a system that still matches the drawings.
What to freeze before the gear leaves the plant
Before factory release, Cx and the owner’s team should lock facts that travel with the unit. If these are fuzzy in the report, fix them at the plant.
Nameplate identity — shop order, serial, kVA, voltages, BIL, cooling class, connection, and the full tap chart.
Governing ratings for the loss / impedance chapter — tap used, reference temperature, and any customer-specific tolerances.
Dielectric history — applied and induced results (and impulse if run), including pass criteria notes.
Liquid condition (liquid-filled) — oil power factor, dielectric breakdown, and any PCB statement.
Open exceptions — anything accepted with note, clouded, or deferred to field verification.
Accessory list vs the one-line — fans, gauges, CT/PT circuits, and similar hardware the SLD assumes are furnished. Factory electrical tests prove the transformer; they do not invent missing tags on project drawings.
The handoff: receiving through pre-energization
Field commissioning should treat the factory package as a serial-number baseline, not a substitute for site tests.
Receive and inspect
Match serial and nameplate to the certified report. Photograph nameplates, shipping damage, nitrogen / blanket pressure if applicable, bushings, and accessory completeness. If the crate and the report disagree on serial or taps, stop—do not sort it out later.
Storage and environment
Insulation power factor and insulation resistance are sensitive to moisture and time. Re-baseline IR and power factor as the project specification requires before trusting the factory dielectric chapter alone.
Installation and terminations
Ratio and polarity were proven on the manufacturer’s bench. Field work still has to prove this HV and LV set of terminations, CT/PT wiring, grounding, and neutral treatment. The factory package will not catch a landed cable on the wrong bushing.
System context
Factory impedance and losses describe the unit. Upstream available fault current, downstream coordination, grounding, and paralleling constraints describe the system. Independent Cx connects those dots before anyone proposes a close-in.
System accessories and drawing reality
Battery disconnects, SPDs, house panels, and similar accessories often live on the project one-line and equipment lists—not as line items inside a transformer factory report. If the SLD never named a battery disconnect, the certified test report will not fix that gap. Cx should flag missing tags, leftover balloons, and sheet-to-sheet fights before buyout and before energization—not after the gear is hot.
Pre-energization hold points
Typical hold points include megger / IR where required, ratio and polarity checks after landing, protective relay settings versus the as-built one-line, interlocking and grounding verification, and a clear go / no-go from the independent commissioning authority—not only from the installer or the manufacturer’s field tech.
How field Cx uses the factory numbers
Good Cx does not re-run every factory test for sport. It uses factory data to ask better field questions:
Losses and impedance — Do short-circuit studies and voltage-drop calculations still match the as-tested %Z on the installed tap?
Ratio and taps — Is the installed tap the one assumed in the voltage study and load schedule?
Dielectric pass — After shipping and landing, do site IR / power-factor trends still support energization?
Identity — Is every downstream document keyed to the same serial and shop order that left the plant?
System integration — Were accessories assumed on the SLD actually furnished, tagged, and landed the way the drawings claim?
Factory electrical tests prove manufacturing quality for a defined test menu. They do not replace independent review of the project documents that tell the field how the unit must live in the system.
What ready to energize should mean
Energization is a system decision. A defensible Cx stance usually requires:
Factory identity and Accepted summary tied to this serial number
Receiving inspection with no open shipping or handling damage
Site verification appropriate to the specification (IR, ratio/polarity after terminations, protection settings)
Alignment between the as-built one-line and what is actually landed in the yard
Open drawing or accessory gaps either closed by the engineer of record or consciously accepted with a documented risk path
Clear roles: manufacturer, installer, and independent commissioning authority
When those pieces line up, closing the breaker is a controlled step—not a leap of faith from a PDF stamped Accepted.
A short owner checklist
Do we have the full certified package for the exact serial on the pad?
Which tap and reference temperature govern the loss / impedance chapter?
What site tests are required after landing, and who signs the hold point?
Do IR / power-factor trends after storage still support dielectric confidence?
Do the as-built one-line and protection settings match the installed unit?
Are there accessory or tagging gaps on the drawings that buyout already skipped?
If any answer is soft, energization should wait.
Closing
Factory witness testing proves the manufacturer built and tested a unit to a standard. Site energization proves the project installed and integrated that unit into a safe, coordinated electrical system. Owners get the full value of both only when independent Cx owns the handoff—from the plant floor to the pad.
If you are planning electrical commissioning for mission-critical or industrial power transformers in the Pensacola area, QEI can help structure factory witness support through field energization—so the report in the crate and the tests on the pad tell one continuous story.
Quinn Electrical and Instrumentation Inc. (QEI) — Pensacola area · Independent electrical commissioning & instrumentation · Info@qei.cx · qei.cx
Educational post. Always follow the project specification, AHJ requirements, and the governing IEEE / NETA (or project-specified) procedures for the equipment class. Switchgear, UPS, and other FAT packages are related but not identical to the transformer suite described here.
Comments