Request a Quote
Share your product list, technical ratings, quantity and delivery destination.
Learn how to read transformer oil DGA reports, compare methods and trends, and prepare the records needed for qualified technical review.
DGA of transformer oil measures gases dissolved in the insulating liquid. Thermal and electrical stresses can generate these gases as oil and insulation materials decompose.[1] Their concentrations and changes over time provide information for assessing transformer condition.[1]
For a buyer or equipment owner, the useful first question is whether the report belongs to the right asset and can be compared with its history. A highlighted result needs context, reliable measurements and a qualified interpretation. IEEE’s public overview of its DGA interpretation guide explicitly includes data quality, reliability and application context.[2]
Laboratory DGA separates and measures gases in an oil sample.[1] ASTM D3612 covers gas extraction and analysis by gas chromatography; its public scope identifies gases including hydrogen, methane, ethane, ethylene, acetylene, carbon monoxide and carbon dioxide, as well as oxygen and nitrogen.[1]
The gas results are evidence to interpret together. ASTM describes their nature, quantity and generation rate as potentially useful indicators of abnormal conditions.[1] It does not describe one gas concentration as a complete assessment of a transformer.[1]
Keep three parts of the document separate: measured results, the laboratory’s interpretation and the asset owner’s action decision. Ask the reviewer to identify the evidence behind a recommendation rather than copying a color or condition label into a purchase decision.

If the equipment description is unfamiliar, first use the oil-immersed transformer guide to understand the tank, insulation and cooling arrangement.
A report is useful only when you know which equipment and sample it describes. Use the following document checks before comparing results. ASTM D923 includes specific techniques for DGA sampling, while Vaisala identifies sample quality and analytical methods as sources of uncertainty.[3][4]
| Report field | What to confirm | Question to resolve before comparison |
|---|---|---|
| Asset and sample identity | Serial number, asset tag, sample location and sample ID | Does this result belong to this transformer and the same oil compartment? |
| Dates | Sample collection, laboratory receipt and analysis dates | Are the historical results arranged by sample date? |
| Insulating liquid | Declared fluid type and any replacement or retrofill history | Is the interpretation method appropriate for that liquid? |
| Analytical method | Named standard, edition and laboratory method | Did the method or laboratory change between reports? |
| Units and reporting basis | Gas units, reference conditions and any conversion notes | Are the concentrations directly comparable? |
| Result qualifiers | Reporting limits, uncertainty and comments on sample condition | What do symbols, blanks and “not detected” mean? |
Keep the original report beside any spreadsheet summary. Preserve qualifiers instead of turning a blank or “not detected” entry into a numerical zero. Ask the laboratory for its definitions before importing the results into a trend chart.
Method names alone do not settle comparability. Vaisala notes differences between ASTM and IEC gas-volume reporting conventions and explains that sampling, extraction, calibration and laboratory processes contribute uncertainty.[3] Request a common reporting basis from the laboratory; avoid applying a conversion copied from an unrelated example.
ASTM identifies both changes in gas concentration and the rate of gas generation as relevant to condition evaluation.[1] A single laboratory sample represents one sampling occasion; it does not supply a continuous record of everything that happened between samples.[6]
Start by placing the available reports in sampling-date order. Keep the intervals visible: two results a few days apart and two results a year apart are different records. Ask the technical reviewer which changes are meaningful given the method, uncertainty and operating context, rather than deciding from the steepness of a spreadsheet line.[2][3]
Attach a short event history to the results. Include the dates of oil treatment, oil replacement, repairs, outages and return to service, together with available load and temperature records. These are review inputs, not proof that an event explains a gas result. Ask the reviewer whether the earlier and later measurements form a valid comparison and which records are missing.

For a transformer offered with only one report, request the earlier reports and maintenance history. If they are unavailable, state that limitation in the purchasing file. A detailed-looking table cannot supply missing history.
A report may contain a gas-ratio interpretation, a Duval chart or a condition status. IEEE’s tutorial abstract describes these within an interpretation process that also addresses data quality, context and selection of reference values.[2] Read the underlying inputs and the method named on the report before treating the label as a conclusion.
The following distinctions help keep a commercial review within its proper scope.
| What you see | What it contributes | What still needs review |
|---|---|---|
| Individual gas concentration | A reported measurement for that sample | Reliability, history and the wider gas pattern |
| Gas trend | A sequence of results over recorded intervals | Comparability and events between samples |
| Ratio or diagnostic chart | An interpretation using a particular method | Applicability, inputs and stated limitations |
| Condition category or alert | A classification under the report’s chosen criteria | The criteria, their edition and relevance to this asset |
| Recommended follow-up | A proposed next step | Responsible engineer, timing and site-specific decision process |
Confirm the fluid type before accepting a familiar interpretation method. Vaisala explains that natural-ester DGA differs from mineral-oil interpretation because the liquids’ chemical compositions affect gas ratios.[5] A mineral-oil reference should not simply be carried across to an ester-filled transformer without technical confirmation.
An online monitor and a laboratory report do not automatically provide interchangeable measurements. Vaisala’s discussion of DGA uncertainty identifies sampling quality, the laboratory process and the monitor’s measurement characteristics as factors in comparison.[3]
Before deciding which result to trust, ask for the matching timestamps, gas channels, units, reporting basis and relevant calibration or uncertainty information. Keep the laboratory’s sample comments with the dataset. Where a comparison remains unresolved, record it as an open technical question rather than averaging the numbers into an apparently tidy answer.
Responsibility for assessing the equipment and deciding its operating status remains with the owner’s qualified team.
Send the complete evidence together, rather than a cropped screenshot of the highest value:
Ask the reviewer to distinguish observations from interpretation, identify missing evidence and state the basis of any requested follow-up. Keep that response with the original reports so that the commercial team can see what was assessed and what remains unresolved.
Do not let an in-service DGA report substitute for the agreed factory acceptance documents. For that separate purchasing task, see the transformer testing guide and define the required test package in the contract.
For an equipment enquiry, send HJD the transformer requirements and available document list. Confirm the offered equipment and documentation scope directly.
Do not base an operating decision on one isolated value. IEEE’s overview places DGA results within a process that considers data quality, reliability and application context.[2] Refer the complete report and equipment history to the responsible qualified engineer.
No. DGA addresses dissolved gases. ASTM’s sampling practice distinguishes DGA from water and particle analyses, each with its own referenced method.[4] Check the agreed test list rather than assuming a document headed “oil analysis” includes every required test.
Request the complete report, the criteria used for the flag, earlier results and the relevant operating and maintenance records. Ask the laboratory to clarify method and measurement questions, then have the responsible engineer define the review and follow-up. Do not choose an operating response from a generic online threshold.