The Project in Brief

The transformer was a 220 kV power unit, nineteen years in service, scheduled for what the maintenance plan described as a standard OLTC service. Three days planned, six-person team, standard spare kit, specialist vendor pre-identified but not pre-scheduled.

By Day 1 afternoon, we knew the three-day plan wasn't going to hold. By Day 3, we had an accurate picture of every decision that had made the project harder than it needed to be — and every support that had prevented it from being worse.

This is the project story, told as a practical account of what helped, what delayed us, and what we would change.

What We Found on Day 1

The first oil sample came from the correct point — the drain at the bottom of the OLTC diverter compartment. We had a discussion during planning about whether the conservator sample from the last quarterly check was representative enough. It wasn't. The bottom-draw sample showed dielectric breakdown voltage below the specification floor and carbon particulate content that explained why — the arcing at the diverter switch contacts had been generating contamination at a higher rate than the upper-layer sampling history had revealed.

The diverter switch removal and contact measurement confirmed the second major finding: three of the four contact positions were beyond the wear replacement threshold. The fourth was within specification but approaching the limit at a rate that would have taken it past the limit before the next scheduled maintenance window.

The motor drive inspection found the root cause of the elevated wear rate: mechanical play in the drive linkage that was producing timing drift in the switching sequence — extending the arc duration per tap change beyond the design specification. Every operation had been generating slightly more arc energy than the design assumed. Over the months since the previous inspection, that increment had accumulated into the contact wear and oil contamination we were now looking at.

By noon on Day 1, the scope was clear: full diverter switch contact replacement, full oil drain and refill, motor drive linkage replacement, and OSR relay calibration verification. The three-day plan was a five-day job.

What Helped — The Support That Made a Difference

The oil sampling equipment was on site before we started.

We had prepared for the bottom-draw sample with the correct equipment — a sampling kit with the appropriate connection for the drain point at the bottom of the OLTC compartment. This sounds like a minor detail. On the previous project I had been involved with, the sampling equipment had been standard conservator sampling kit — not suitable for a representative bottom-draw. We arrived prepared this time because someone had asked the question during the planning phase: where exactly are we sampling from, and do we have the equipment for that specific point?

The EMR Global spares contact resolved the motor drive component question efficiently.

When the motor drive linkage play was confirmed as exceeding the specification limit, the immediate question was whether the correct replacement components could be identified and sourced within the outage window. We contacted EMR Global's spares support with the motor drive unit type, the OLTC design, and the specific linkage assembly description from the inspection.

The technical identification was confirmed within the hour — the correct linkage components for our specific motor drive configuration, not just the general design family. The expedited supply arrangement was confirmed for delivery within the extended outage timeline. This was the support interaction that had the most direct impact on the project timeline — not because it solved everything, but because it prevented the component identification delay from extending the outage beyond the five days the expanded scope required.

The testing sequence was documented before the outage.

We had prepared the testing sequence during the work package phase — the specific order of steps from contact resistance before oil fill through to OSR relay verification after pipeline reconnection. When the expanded scope arrived on Day 1, the testing sequence structure was already in place. We extended the scope within the existing structure rather than rebuilding the sequence under time pressure.

What Delayed Us — The Gaps We Could See in Retrospect

The Tier 2 spares were not pre-ordered.

The motor drive linkage components were not in the original spare kit because the original scope had not anticipated motor drive linkage replacement. When the Day 1 inspection confirmed the need, the procurement question became: how quickly can these components arrive?

The answer was: faster than if we had called a supplier without an established relationship, but still subject to a delivery window that compressed our timeline. If we had placed a contingency order for motor drive linkage components — on a return-if-unused basis, as part of the pre-outage Tier 2 spare ordering — those components would have been on site on Day 1. The delivery wait would have been zero.

The lesson we took from this: every pre-outage spare conversation should include a Tier 2 run — what are the five components most likely to be needed if the inspection finds worse-than-expected condition? Order those components on contingency terms before the outage opens.

The specialist vendor for OSR relay calibration was not pre-scheduled.

The OSR relay calibration step required specialist equipment and a qualified technician. We had identified the vendor during planning but had not confirmed scheduling — the assumption was that we would call when we reached that step and they would be available.

They were available, as it turned out. But the scheduling confirmation and travel time added a half-day to the project timeline at the point when every half-day was visible on the operations department's daily timeline update. A pre-scheduled availability — with a no-charge cancellation window if we reached the step earlier than planned — would have had the specialist on site within the project sequence rather than waiting for them to arrive.

The operations department was briefed on the base timeline but not the contingency.

When we communicated the expanded scope and the five-day revised timeline on Day 1 afternoon, the operations response was genuine surprise. They had planned load redistribution for three days. Five days required a revised load management plan, additional control centre communication, and an impact assessment on adjacent scheduled work — all of which was handled efficiently but all of which consumed coordination time at exactly the moment the maintenance team needed clear decision space.

If the operations briefing before the outage had included the contingency scenario — specifically, that the outage could extend to five days if Day 1 inspection confirmed the motor drive and contact findings that the condition data hinted at — the extended timeline would have been a pre-planned possibility rather than an unplanned development.

What We Would Change Next Time

Six weeks before the outage:

• Pre-outage condition review including an explicit worst-reasonable-case assessment
• Tier 1 and Tier 2 spare orders placed, both specified against the actual OLTC design
• Specialist vendor availability pre-scheduled with cancellation window
• Operations briefed on both base and contingency timelines

Two weeks before:

• All spare deliveries confirmed
• Testing sequence and documentation forms in the work package
• Procurement team briefed for potential rapid-response additional requirements
• Supply partner contact confirmed for Day 1 rapid response

Day 1 morning:

• Bottom-draw oil sample as first action after OLTC compartment access
• Day 1 inspection findings documented on structured forms in real time
• Scope confirmation or expansion decision communicated to all stakeholders by midday
• Any Tier 2 spare requirements communicated to procurement before 2 PM

What stays the same:

The technical work was done correctly. The contacts were replaced with specification-correct components. The oil was tested before fill and confirmed within specification. The OSR relay was calibrated to the correct threshold. The motor drive timing was verified through the complete switching sequence before the compartment was sealed. The return-to-service testing sequence was completed in the correct order and documented before isolation was removed.

The technical execution was right. The planning execution had gaps. The gaps were visible in the timeline. Next time, the planning execution will be as good as the technical execution — and the project will run five days because it needed to be five days, not because it ran into delays that earlier planning would have prevented.