The Key Strategic Decision
After deciding to implement online DGA monitoring, the next critical question is: single-gas (hydrogen-only) or multi-gas? This decision has profound implications for fault detection capability, diagnostic depth, and total cost.
Key figures at a glance
- Multi-gas monitors measure on a cycle of 1 to 24 hours (default 4 hours) — up to 6 measurements per day with no operator involvement.
- Palladium-alloy hydrogen sensors hold ±15% of reading or ±5 ppm accuracy with no routine calibration.
- The laser photoacoustic (LPAS) detection cell has no consumables and its laser diode is rated for more than 50,000 hours of continuous operation.
Hydrogen-Only Monitoring
What it measures: H₂ (hydrogen) only, sometimes with moisture.
How it works: A solid-state sensor (palladium thin-film, thermal conductivity, or fuel cell) in direct contact with the oil or headspace gas detects hydrogen concentration.
Typical products: Siemens H2Guard Lite, Hitachi CoreSense, Qualitrol TM1/DGA 150, Vaisala MHT410
| Advantage | Limitation |
|---|---|
| Lowest cost ($2,000–$8,000 per unit) | Cannot identify fault type — H₂ is a universal indicator |
| Small, simple installation | Cannot distinguish PD from thermal fault from arcing |
| Very low maintenance | Cannot detect acetylene ≤2 ppm (arcing!) |
| Suitable for distribution transformers | No Duval Triangle/Pentagon diagnostics possible |
| Good first-line screening | Rising H₂ requires lab GC follow-up to diagnose |
Hydrogen-only is appropriate for:
- Distribution transformers (<10 MVA) — fleet-wide early warning
- Transformers with very low failure consequence
- First-stage screening — identify which transformers need multi-gas
- Budget-constrained programs where any monitoring is better than none
The critical limitation: Acetylene can be as low as 2 ppm during an active arcing fault. A hydrogen-only monitor cannot detect this — and H₂ may not rise significantly in some arcing scenarios until the fault is already catastrophic.
Multi-Gas DGA Monitoring
What it measures: 5–9 gases: H₂, CH₄, C₂H₆, C₂H₄, C₂H₂, CO, CO₂ + moisture (and optionally O₂, N₂).
How it works: Photoacoustic spectroscopy (PAS), gas chromatography (GC), NDIR, or FTIR analyzes extracted gas for all fault gases simultaneously.
Typical products: GE Kelman DGA 900, PAS DGA DGA-200/500, Vaisala OPT100, Qualitrol TM8
| Advantage | Limitation |
|---|---|
| Full fault diagnosis — Duval, IEC, Rogers methods | Higher capital cost ($15,000–$50,000) |
| Detects acetylene — the most critical fault gas | Larger physical footprint |
| Distinguishes thermal faults from electrical faults | Requires oil extraction system |
| CO/CO₂ for paper degradation assessment | |
| Rate-of-change trending for all gases | |
| Standards-compliant diagnostics (IEC/IEEE) |
Multi-gas is recommended for:
- All transformers ≥100 MVA or ≥220 kV
- GSU transformers (any size)
- HVDC converter transformers
- Critical industrial transformers
- Any transformer where fault diagnosis is required (not just detection)
What Hydrogen-Only Monitoring Misses
Consider these real fault scenarios where H₂ monitoring alone would fail:
Scenario 1: Developing Arcing Fault
- C₂H₂ rises from 0 to 5 ppm over 3 days
- H₂ remains within normal range (60 ppm → 85 ppm — still below 90 ppm IEEE norm)
- H₂-only monitor: No alarm triggered
- Multi-gas monitor: Immediate C₂H₂ alarm → emergency response
- Outcome without multi-gas: Catastrophic failure within 1–2 weeks
Scenario 2: Paper Insulation Degradation
- CO rises from 300 to 800 ppm over 18 months
- H₂ remains stable at 40 ppm
- H₂-only monitor: Normal reading — no indication of problem
- Multi-gas monitor: CO/CO₂ ratio trending down → cellulose pyrolysis alert
- Outcome without multi-gas: Insulation failure discovered at next outage; complete rewind required
Scenario 3: Thermal Fault Progression (T1→T2→T3)
- CH₄ and C₂H₄ steadily increasing over 6 months; C₂H₄/C₂H₆ ratio crossing from <1.0 to >4.0
- H₂ moderately elevated but stable at 120 ppm
- H₂-only monitor: Flagged as “elevated hydrogen” — scheduled for lab follow-up in 3 months
- Multi-gas monitor: Clear T2→T3 thermal fault progression → planned outage
- Outcome without multi-gas: Fault reaches T3 severity; carbonized oil requires complete oil replacement + extensive internal inspection
The 3-Gas Middle Ground
An emerging compromise: 3-gas monitoring (H₂ + C₂H₂ + moisture or H₂ + C₂H₂ + CO).
| Product | Gases | Technology |
|---|---|---|
| Qualitrol TM3 | CH₄, C₂H₄, C₂H₂ | GC |
| GE Kelman MINITRANS | H₂, C₂H₂, CO | PAS |
| Weidmann InsuLogix HMA | H₂, C₂H₂, moisture | Laser + TC |
Advantages:
- Detects acetylene (the most critical single gas beyond H₂)
- Lower cost than full 9-gas
- Simpler than full multi-gas
Limitations:
- Cannot perform full Duval Triangle diagnostics (requires CH₄, C₂H₄, C₂H₂)
- Limited thermal fault assessment without C₂H₄
- Still requires lab GC for complete diagnosis
The 3-gas approach is gaining traction for medium-voltage transformers (69–138 kV) where full 9-gas may be cost-prohibitive but H₂-only provides insufficient protection.
Decision Matrix
| Transformer Type | Voltage | MVA | Recommended | Reason |
|---|---|---|---|---|
| GSU (nuclear/thermal) | Any | >100 | 9-gas PAS or GC | Maximum protection; regulatory |
| GSU (renewable) | Any | 50–300 | 9-gas PAS | Remote; zero consumables |
| Transmission | ≥220 kV | >100 | 9-gas PAS | Full diagnostics; unmanned |
| Transmission | 69–138 kV | 20–100 | 5–9 gas PAS | Good diagnostics |
| Sub-transmission | 33–69 kV | 10–50 | 3–5 gas or H₂+C₂H₂ | Cost-effective protection |
| Distribution | <33 kV | <10 | H₂-only or 3-gas | Fleet screening |
| Industrial critical | Any | Any | 9-gas PAS | Downtime cost driven |
| Offshore/marine | Any | Any | 9-gas PAS | Zero access; unmanned |
| OLTC | — | — | Dedicated 3–5 gas | Separate oil compartment |
The Real Cost Argument
If a hydrogen-only monitor costs $5,000 and a multi-gas PAS monitor costs $25,000, the $20,000 difference seems significant. But consider:
| Cost Element | Amount |
|---|---|
| Transformer replacement (100 MVA) | $1,500,000 |
| One week generation outage (100 MW @ $50/MWh) | $84,000 |
| Emergency crew mobilization | $50,000 |
| Environmental cleanup (oil spill) | $100,000+ |
| Regulatory penalties | Variable |
| Total potential failure cost | $1,734,000+ |
The $20,000 incremental cost for multi-gas represents 1.2% of a single avoided failure — and multi-gas monitoring provides the diagnostic capability to identify and locate faults before they become failures, something hydrogen-only monitoring cannot do.
Summary
| Hydrogen-Only | Multi-Gas PAS | |
|---|---|---|
| Detects a problem? | ✅ Yes | ✅ Yes |
| Identifies WHAT problem? | ❌ No | ✅ Yes — Duval, IEC, Rogers |
| Detects acetylene (arcing)? | ❌ No | ✅ Yes — down to 0.1 ppm |
| Tracks paper degradation? | ❌ No | ✅ Yes — CO/CO₂ ratios |
| Guides maintenance decisions? | ⚠️ Limited | ✅ Yes — fault-specific |
| Meets IEEE C57.143 recommendation? | ❌ No | ✅ Yes — for critical assets |
| Cost per unit | $2K–$8K | $15K–$50K |
| Best application | Fleet screening, distribution | Critical assets, transmission, GSU |
Next Steps
- DGA Fundamentals — Understanding what each gas tells you
- Online vs Offline DGA — Building your monitoring strategy
- Product Selector — Choose the right DGA monitor for your assets
How to decide between a hydrogen-only and a multi-gas monitor
- Step 1: Define the fault coverage you need — hydrogen is a universal early-warning gas present in about 90% of developing faults; multi-gas adds full fault-type diagnosis.
- Step 2: Assess asset criticality — for UHV/HVDC or generator step-up transformers choose multi-gas; for general distribution, hydrogen-only is often enough.
- Step 3: Compare cost and form factor — hydrogen sensors (DGA-200/300/500) are smaller and lower-cost; a multi-gas system (DGA-900) measures 9 gases plus moisture.
- Step 4: Check maintenance burden — palladium-alloy hydrogen sensors need no routine calibration, and the LPAS cell has no consumables.
- Step 5: Select the output and integrate — confirm the monitor supports your SCADA protocol before purchase.
FAQ
Q: What is the difference between hydrogen-only and multi-gas monitoring?
A: Hydrogen-only monitors (DGA-200/300/500) detect H₂ as a universal early-warning gas — they catch ~90% of developing faults at lower cost. Multi-gas monitors (DGA-900) measure all 7+ fault gases, enabling full fault-type identification via Duval Triangle, IEC ratios, and Rogers Ratio — they tell you not just that there is a fault, but what kind and where.
Q: What gases does a DGA monitor detect?
A: The DGA-900 detects 9 fault gases plus moisture: hydrogen (H₂), carbon monoxide (CO), carbon dioxide (CO₂), methane (CH₄), ethane (C₂H₆), ethylene (C₂H₄), acetylene (C₂H₂), oxygen (O₂), nitrogen (N₂), and moisture (H₂O). The DGA-200/300/500 hydrogen sensors detect H₂ down to 1–2 ppm.