Transformer SFRA Baseline for EPC Handover: Configuration, Comparison and Records

A useful transformer SFRA baseline is more than a set of curves attached to a factory report. Zisheng Electric recommends treating it as a controlled engineering record that the site team can understand, reproduce and hand over to the owner. If the original measurement configuration cannot be reconstructed, a later comparison may create uncertainty at exactly the point when the project needs a clear decision.

For an EPC contractor, the practical question is whether the factory and site records support the same comparison. The answer depends on the equipment identity, test configuration, usable data files and the review process agreed before delivery. This guide focuses on those interfaces for oil-immersed power transformers. It is intended for procurement engineers, inspectors, commissioning managers and owners preparing a diagnostic handover package.

What a transformer SFRA baseline can establish

Sweep frequency response analysis measures a transformer’s electrical response across a range of frequencies. It is used to investigate the mechanical and electrical condition of its active part, including after transport or high-current events. The comparison requires a reproducible measurement arrangement; the OMICRON explanation of SFRA measurement and connection repeatability describes why this matters.

The official scope of IEC 60076-18:2012 addresses the measurement technique and equipment for frequency response measurements in the factory or at site. A procurement specification should identify its adopted edition and the applicable project procedure. Referencing the standard does not, by itself, define the project’s document delivery dates, review authority or response to an unexplained difference.

Keep those commercial and engineering responsibilities explicit. The supplier should know what evidence is required before shipment. The site contractor should know which files and accessories must be available before the test team mobilizes. The owner should know who is authorized to accept the completed diagnostic review.

Start with the comparison the project will need

Separate factory, transport and installation records

Use distinct record sets for distinct equipment conditions. A factory reference, a transport-condition measurement and an installed reference should have separate identifiers whenever their configurations differ. Record the purpose of each set in plain English. Calling every file “baseline” makes it difficult to decide which one should be used for a later comparison.

Plan the sequence against the actual delivery scope. Ask whether service bushings remain installed during shipment, whether the transformer travels filled or in another agreed condition, and when final assembly will be complete. The receiving inspection schedule should identify the measurement stage and the person responsible for confirming readiness. Avoid leaving that decision to whoever happens to be available on delivery day.

Write the review question before booking the test

A useful request might be: “Compare the agreed site measurement set with the identified factory reference and explain any material difference after checking configuration compatibility.” That wording asks for a reasoned review. It also makes the deliverable broader than a technician’s attendance sheet or a screenshot showing that a measurement was completed.

Include the SFRA record in the project’s transformer FAT witness plan, with a stated document review point and a named recipient. The witness team can then resolve missing information while the transformer and the factory test personnel are still available.

Build a configuration record that another team can use

Technical literature identifies terminal conditions, measurement direction, source settings, connection arrangements, tap position, bushings, insulating fluid and residual magnetization as factors that can affect comparison. The paper by Uhrig and Rädler on reproducible frequency response measurements also explains why a difference between traces is not automatically evidence of deformation. Use that distinction when designing the record sheet.

The table below is a proposed EPC document-control aid. It does not replace the approved test method or specify electrical connection instructions. Its purpose is to show which missing records should trigger a question before the comparison is accepted.

Record item	Evidence to retain	Decision supported	Action if incomplete
Equipment identity	Serial number, unit designation and applicable nameplate record	Confirm that both datasets belong to the intended transformer	Resolve identity before comparing files
Measurement configuration	Terminal schedule, measurement direction and approved connection sheet	Establish whether the test arrangements are comparable	Obtain the original setup record or qualify the comparison
Physical condition	Bushing arrangement, fluid condition and assembly-stage record	Select the appropriate reference dataset	Refer differences to the diagnostic reviewer
Tap and test history	Recorded tap position and relevant preceding test activities	Check whether the intended reference condition was reproduced	Clarify the sequence before interpreting a deviation
Instrument and accessories	Equipment identifier, verification record and setup photographs	Assess measurement traceability and repeatability	Document checks and determine whether a repeat is needed
Results and review	Raw files, readable report, comparison notes and signed disposition	Close the diagnostic review with an auditable rationale	Keep the review item open until the evidence is usable

Assign one owner to each missing item. “Awaiting clarification” is not a useful close-out description unless the request, recipient and required response are visible. A short register with accountable actions is more useful than a large folder of unindexed attachments.

Photograph the setup, then protect the original data

Capture information that a curve cannot show

Ask for an overall view of the transformer and separate photographs that identify the measurement arrangement. Include enough context to relate each photograph to the relevant connection sheet. A close-up with no recognizable equipment reference may be difficult to interpret months later. Photograph names should connect to the test record rather than remain as an unexplained sequence from a camera.

Prepare a simple naming convention before FAT. For example, combine the project unit designation, measurement stage, record identifier and revision. The exact syntax is less important than consistent use. The same identifier should appear in the report index, setup photographs and transmitted data folder, so a reviewer does not have to infer which items belong together.

Deliver files the owner can reopen

Specify raw measurement files and a readable report as separate deliverables. Ask the supplier to identify the software or viewer needed to inspect the original data, together with any export format available. Have the receiving team open a sample file before accepting the final package. Discovering a missing viewer after demobilization can delay an otherwise straightforward review.

Keep the original measurements read-only in the project archive. Store working comparisons separately and record which originals were used. If a report is corrected, retain the earlier issue with its revision history instead of silently replacing it. This protects traceability when different parties have downloaded different versions.

Resolve comparability before discussing transformer condition

When a site result differs from the reference, first check that the correct records were selected. Confirm the transformer identity and compare the configuration sheets line by line. Where the records leave a gap, record the gap as an evidence problem. Do not turn missing information into a confident diagnosis simply because the programme needs an answer.

Use a three-stage review note: observation, explanation and disposition. The observation describes the difference without naming a fault. The explanation records the evidence considered and any remaining uncertainty. The disposition states whether the review is complete, whether more information is needed, or whether further investigation has been agreed.

Where repeat measurement is appropriate, the approved procedure should define who authorizes it and which conditions must be documented. Retain both the original and repeat files. A better-looking repeat trace should not erase the reason the repeat was requested. The reviewer needs the sequence to understand what changed.

Do not insert a universal percentage difference or correlation threshold into a purchase order without an agreed technical basis. If the project uses an assessment index, require its definition, applicability and review method. A single acceptance number should not conceal questions about the equipment configuration or the quality of the source record.

Connect the review to delivery decisions

SFRA records should sit alongside the receiving inspection, assembly records and transformer transport shock monitoring file. Keep their purposes distinct. A logistics record describes the shipment history; the diagnostic review evaluates the measurements available. The project should bring these records together through a documented engineering decision.

Define the authority for shipping release, installation release and energization release separately. Closing an SFRA document comment does not automatically close every other inspection or commissioning item. Conversely, an administrative omission should be described precisely so that the team can distinguish missing paperwork from an unresolved technical concern.

If the diagnostic reviewer requests additional checks, record the scope, responsible organization and effect on the programme. Tell the procurement team whether the action requires factory support, site access, specialist attendance or additional documents. This allows resources to be arranged around a defined task rather than a vague instruction to “check the transformer.”

Use a handover exercise before the test team leaves

Choose one measurement set and ask a receiving engineer who did not perform the test to locate its raw file, report, configuration sheet and photographs. Have that engineer explain which transformer and measurement stage the set represents. This is a practical document usability check, not a substitute for technical interpretation.

Then verify that the approved comparison report can be connected to its source files. Record the reviewer, review date and outstanding limitations. If a correction is requested, define the exact deliverable and due date. Close the comment only after the revised evidence has been received and checked.

The final archive should remain usable after the project team changes. Give the owner an index, the original data, the approved reports and the contact route for technical questions. Confirm the handover location and receipt through the project’s normal document system. An email saying “files attached” is weak evidence if the attachments are incomplete or their relationship is unclear.

Questions EPC buyers should settle before ordering

What if the factory baseline is missing?

Request a written assessment of what comparison is possible with the records available. Other reference approaches may require a different level of specialist judgment. Do not relabel a measurement from another unit as this transformer’s original factory baseline. Describe the limitation and agree the next step with the responsible engineer.

Who should own the comparison report?

The purchase order and commissioning scope should identify the party responsible for preparing it and the party responsible for acceptance. They may be different organizations. Specify how the transformer supplier will support technical queries and how review comments will be consolidated, so conflicting instructions do not reach the test team.

Must the same company perform both measurements?

The more useful procurement question is whether both teams can work to the agreed procedure and provide comparable, traceable records. If different contractors are appointed, exchange the proposed record template and file requirements before mobilization. Resolve interface gaps while changes are inexpensive.

Prepare the transformer SFRA baseline scope with Zisheng Electric

For projects considering Zisheng Electric’s 110 kV and 115 kV power transformers or 133 kV and 135 kV three-phase power transformers, include the required diagnostic records in the technical inquiry. The actual test scope and delivery configuration should be agreed for the selected equipment and contract.

Send the drawings, datasheets, load list, technical specifications and single-line diagram, together with grid parameters, environmental conditions and installation-site conditions. Add the proposed factory and site measurement stages, document format requirements and review responsibilities. These inputs help define a transformer SFRA baseline package that supports the project’s delivery decisions.

Our engineering team will review the requirements and respond to project inquiries within 24 hours.