DGA analyzer
Interactive DGA Analyzer
Enter seven dissolved-gas concentrations in ppm and see the Duval Triangle 1 placement, the IEC 60599 basic gas ratios, and a screening against IEEE C57.104-2019 typical values — computed entirely in your browser.
Interactive DGA Analyzer
Enter seven dissolved-gas concentrations in ppm and see the Duval Triangle 1 placement, the IEC 60599 basic gas ratios, and a screening against IEEE C57.104-2019 typical values — computed entirely in your browser.
Runs entirely in your browser — the values you enter never leave this device.
Duval Triangle 1
Relative shares of CH4, C2H4 and C2H2, plotted against the IEC 60599 fault zones.
- PD
- Partial discharges
- T1
- Thermal fault, below 300 °C
- T2
- Thermal fault, 300 °C to 700 °C
- T3
- Thermal fault, above 700 °C
- DT
- Mixed thermal and discharge fault
- D2
- Discharges of high energy
- D1
- Discharges of low energy
Indicated zone
No sample plotted yet — the diagram shows the published fault-zone map.
- % CH4
- —
- % C2H4
- —
- % C2H2
- —
Basic gas ratios
The three basic ratios used by IEC 60599 and the Rogers method for ratio-based screening.
- C2H2 / C2H4
- —
- CH4 / H2
- —
- C2H4 / C2H6
- —
Screening against IEEE C57.104-2019 typical values
Each entered gas compared with the published 90th-percentile typical values (IEEE C57.104-2019, Table 1, O2/N2 ratio ≤ 0.2). Typical values are statistical reference thresholds — not limits, and not a diagnosis.
| Gas | Your value | Typical value (90th percentile) | Screening |
|---|---|---|---|
| H2 | — | 80 | No value yet |
| CH4 | — | 90 | No value yet |
| C2H2 | — | 1 | No value yet |
| C2H4 | — | 50 | No value yet |
| C2H6 | — | 90 | No value yet |
| CO | — | 900 | No value yet |
| CO2 | — | 9,000 | No value yet |
What this tool is — and is not
This is a planning aid computed entirely in your browser. It does not assess transformer condition, does not replace IEEE or IEC guidance or engineering judgment, and its output is not a diagnostic conclusion. Use it to prepare a pilot conversation.
- No diagnostic authority
- The tool does not determine transformer condition, approve maintenance, certify standards compliance, or replace qualified engineering judgment.
- No private source intake
- Do not enter confidential asset names, work-order IDs, network details, customer data, or proprietary measurements into public planning tools.
- Human review stays central
- GridAPM positions AI draft output as review material until authorized reviewers approve, edit, reject, or escalate it.
Related tools
Continue scoping the pilot surface.
These adjacent tools help convert readiness, evidence gaps, governance questions, and handoff fields into a stronger pilot brief.
Pilot Brief Builder
Generate a first-pass GridAPM pilot brief for transformer APM evidence, workflow pain, reviewer roles, data boundaries, and success metrics.
Open tool Work packageTransformer Maintenance Work Package Builder
Draft a human-reviewed transformer maintenance work-package outline from condition evidence, maintenance history, criticality, approval states, and CMMS/EAM handoff fields.
Open tool OT AI riskUtility OT AI Risk Register Builder
Create a first-pass risk register for approval-gated transformer APM AI workflows across OT boundaries, source approvals, agent permissions, cybersecurity review, and audit artifacts.
Open toolFAQ
Boundaries for this tool.
What is dissolved gas analysis (DGA)?
DGA measures the gases dissolved in a transformer's insulating oil. Faults such as overheating, partial discharge, and arcing decompose the oil and the paper insulation into characteristic gases (H2, CH4, C2H2, C2H4, C2H6, CO, CO2), so the gas pattern is a screening signal for incipient faults.
What does the Duval Triangle 1 show?
Duval Triangle 1 plots the relative percentages of CH4, C2H4 and C2H2 in a ternary diagram whose zones — PD, T1, T2, T3, D1, D2 and DT — correspond to the fault types published in IEC 60599. It indicates the likely fault type once a fault is suspected; it does not decide whether a fault exists.
Are the IEEE C57.104-2019 typical values pass/fail limits?
No. They are 90th-percentile statistical values from large transformer populations — a screening reference, not condemnation limits. A value above typical means the gas deserves attention and trending, not that the transformer has failed; interpretation depends on rates of change, history, and context.
Can this analyzer replace a laboratory or an engineer?
No. It is an educational screening aid that runs deterministic, published calculations in your browser. Real assessments rely on accredited laboratory measurements, gas trends over time, and review by a qualified engineer — this tool supports learning and first-pass screening only.
Turn the result into a pilot conversation.
Share the non-confidential result with GridAPM to discuss approved evidence sources, security boundaries, reviewer ownership, and a measurable first workflow.