Condition-Based Replacement of a Critical 138/69 kV Transformer
A utility owns 12 138/69 kV transformers. Unit T-7 is 45 MVA, 48 years old, and carries 38 MW at summer peak, about 90% of its forced-cooled rating. Dissolved gas analysis shows acetylene rising from 2 to 12 ppm in six months, ethylene 85 ppm, and moisture 35 ppm; oil tests show interfacial tension falling. A thermal scan shows a 12 degrees C rise on one radiator bank, and a recent load tap changer inspection found pitting. The fleet spare is a 30 MVA mobile transformer that can be connected in 72 hours but cannot carry the full load; a 30-month lead time applies to a new 45 MVA unit costing $6.5 million. The three-year capital budget is $10 million, and another 138/69 kV unit, T-11, has a bushing with a known defect and a 9-month replacement lead time. The utility has a NERC TPL-001 requirement to avoid uncontrolled loss of load for single contingencies and a reliability standard that treats transformer failure as a credible event.
1. Rank T-7 and T-11 for replacement and explain.
Read the worked answer
Use risk = probability x consequence. T-7 has active DGA evidence of arcing, with acetylene at 12 ppm and rising, plus moisture, a hot radiator, and LTC pitting; it carries 90% load and has no full-size spare. Probability of failure is elevated and consequence is loss of load and possible collateral damage. T-11 has a known bushing defect but a 9-month lead time and presumably lower probability until testing shows otherwise. Both are credible single contingencies under TPL-001. Rank T-7 first for replacement and immediate operating restrictions; order both the 45 MVA transformer and the T-11 bushing. If only one capital project can start, T-7; but the bushing is a low-cost, high-value fix and should not wait for the transformer.
2. What operating limits and monitoring would you impose on T-7 until replacement?
Read the worked answer
Limit loading to 80% of forced-cooled rating unless a real-time contingency analysis justifies higher; transfer load to adjacent transformers; verify the 30 MVA mobile can carry critical load and pre-position cables. Increase DGA to weekly and add online gas monitoring for acetylene and hydrogen; trend moisture and top-oil and winding temperatures. Inspect the LTC and radiator; if acetylene rises above the IEEE C57.104 condition threshold or moisture exceeds limits, remove from service. Prepare a contingency plan for a failed transformer: switching, mobile connection, load-shed priorities, and customer notification. Do not rely on the 72-hour mobile connection for full load.
3. The budget is $10 million for three years. How would you spend it, and what would you tell the commission?
Read the worked answer
Allocate $6.5 million for the 45 MVA transformer with early procurement to hold the 30-month lead time, $0.5 to $1.0 million for the T-11 bushing and testing, and the remainder for monitoring, mobile-cable upgrades, and a spare-strategy study. Refurbishing T-7 is not justified when acetylene indicates arcing; refurbishment may not remove the fault and can cost 40 to 50% of new. Deferring the transformer is a false economy because the expected outage cost and safety risk exceed the carrying cost. Tell the commission that the project is condition-based and reliability-driven, show the DGA trend, load at risk, and the consequence of failure. If the budget is capped, request a contingency or lease a spare; do not silently accept the risk.