Build, Maintain, Upgrade: A lifecycle approach to Critical Power Infrastructure
Updated: Sep 22
──── FEATURED ARTICLES
Build, Maintain, Upgrade: A lifecycle approach to Critical Power Infrastructure
Most critical power infrastructure is bought the way it is budgeted: as a series of discrete events. A new build is one procurement. Maintenance is a recurring line item, often competitively bid each cycle. A modernization or retrofit is a separate capital project years later. Each decision is rational on its own; together, they quietly raise operational risk and total cost of ownership across the life of the facility.
The hidden cost of treating the lifecycle as three projects
The expense of a fragmented lifecycle rarely shows up as a single large invoice. It shows up as friction, distributed across years. When a new maintenance vendor inherits a plant they did not build, the first months are spent rediscovering the system: which trip units were set to what, how the transfer schemes were sequenced at energization, and where the as-built drawings diverge from what is installed. At upgrade time the pattern repeats — a retrofit team with no knowledge of the plant’s quirks re-surveys everything from drawings that may be a decade out of date.
Industry analyses from groups such as the Uptime Institute have long observed that a large share of data center outages trace back not to catastrophic equipment failure but to human error and procedural gaps — exactly the kind of organizational gap that fragmented accountability produces. With a single mission-critical incident commonly running into significant direct costs and even greater downtime costs, that include recovery, and contractual exposure — the cost of lost institutional memory is not abstract. A lifecycle approach attacks this directly: nothing has to be rediscovered, because the record was never handed off to a party that did not already hold it.
Build: A baseline designed to be maintained
The lifecycle begins before the first piece of equipment is energized. How the system is specified, coordinated, and documented during a new build determines how maintainable and upgradable the facility will be for its entire operating life. A systems integrator working across the full lifecycle approaches this differently from a vendor whose involvement ends at handover: because the same team expects to maintain and eventually modernize the plant, as-built drawings, protective device settings, and coordination studies are captured as a living record that serves operations rather than a binder on a shelf. This is also where vendor-neutral integration earns its place — a facility assembled from the best-fit components for each subsystem, rather than one manufacturer’s catalog end to end, is more serviceable over twenty years because the operator is not locked into a single supplier’s parts and service channel.
JDC supports the owner’s overall design with component, section, and system-level coordination across multiple OEM partners, then carries that discipline into start-up and commissioning support, where the integrated systems testing that verifies source transfers, redundancy paths, and concurrent maintainability also produces the as-installed baseline every future decision references. Importantly, a lifecycle integrator coordinates the power system around — and integrates with — the utility service, standby generators, UPS, PDUs, and electrical power monitoring system (EPMS).
Maintain: from preventative to predictive
Maintenance is where the lifecycle is either protected or eroded, and it is the phase most often treated as an interchangeable commodity. The distinction that matters is between preventative and predictive maintenance. Preventative maintenance is interval-based: equipment is inspected, tested, and serviced on a defined schedule regardless of measured condition. The NETA Maintenance Testing Specifications (MTS) provide widely referenced intervals and acceptance criteria for this work — insulation resistance testing, contact resistance measurements, mechanical operation testing on breakers and switches, protective relay testing, and infrared thermographic scanning of energized equipment to surface loose connections and thermal anomalies before they fail.
Predictive maintenance goes a step further, using condition data to forecast when intervention is actually needed. Partial discharge monitoring on medium voltage switchgear, for example, can detect insulation degradation that thermographic scanning cannot see, while EPMS trending can reveal loads drifting toward limits or breakers operating outside their expected envelope. The objective is to move from “service everything on a calendar” toward “service what the data says needs servicing, before it fails” — without abandoning the interval testing that catches what monitoring misses. A lifecycle partner runs this well precisely because they hold the baseline: predictive maintenance is only as good as the reference point it trends against, and a team that established the as-installed condition at start-up can tell a meaningful change from normal variation far faster than a vendor reading the equipment cold. Nationwide in-house field service teams preserve that continuity across maintenance cycles and across the inevitable turnover in the operator’s own staff.
A critical power lifecycle-maintenance planning checklist
To gauge whether a maintenance program is protecting the lifecycle rather than just satisfying a line item:
1. Is there a documented as-installed baseline? Confirm a commissioning condition record exists for switchgear, switchboards, and protective devices, and that maintenance is trended against it.
2. Are NETA MTS intervals defined and followed? Verify that insulation resistance, contact resistance, mechanical operation, and protective relay testing are scheduled to recognized intervals, not deferred ad hoc.
3. Is thermographic scanning part of the routine? Confirm energized infrared scanning of critical paths runs on a defined cycle and the results are logged.
4. Is any predictive monitoring in place? Assess whether partial discharge monitoring on medium voltage equipment and EPMS trending is used to forecast issues, not just record them.
5. Are protective device settings and arc flash labeling current? Verify the coordination study and arc flash hazard analysis still match the system as installed.
6. Is parts and service availability confirmed? Check that certified replacement parts and trained technicians remain available for the installed equipment and its embedded relays and trip units.
7. Does one party hold the institutional record? Determine whether the team performing maintenance also holds the build documentation and carries it into the next upgrade — or whether the record fragments at every contract boundary.
8. Engage third-party consultants. Having a critical systems consultant, often the installation team EOR and/or CxA evaluate the maintenance program periodically can be the basic sounding board to ensure “group think” does not become an issue. It also keeps these professionals engaged for future modification planning and design.
A program that can answer these affirmatively is maintaining a lifecycle. A program that cannot is maintaining equipment in isolation.
Upgrade: Modernization as a planned phase, not an emergency
Every electrical system eventually reaches the point where maintenance alone cannot keep pace — with aging equipment, obsolete parts, or workload demands the original design never anticipated. Whether that moment arrives as a planned modernization or an emergency replacement depends almost entirely on how well the maintain phase fed into the upgrade decision.
A lifecycle approach turns modernization into a sequencing problem rather than a crisis. Because the maintenance history already documents which equipment is showing thermal anomalies, mechanical wear, or partial discharge activity, and because parts availability has been tracked rather than assumed, the upgrade can be scoped against real condition data instead of a fresh survey.
Switchgear and switchboard modernization can then be planned as distinct workstreams — medium voltage switchgear and the low voltage switchboards inside the building have different lifecycles and failure modes, and a serious plan addresses both.
The defining constraint is that the work has to happen around live load. A phased retrofit while energized depends on the existing protective device coordination, the current redundancy paths, and the maintenance windows the operation can absorb — and concurrent maintainability, the ability to take any element offline for replacement without dropping the load, is what makes that retrofit possible. It is far easier to execute when the team already knows how the plant was built.
Start-up and commissioning support for the new equipment then re-verifies the system and re-establishes the baseline, closing the loop back to where the lifecycle began. Also, keep in mind that when a common phased approach is used for such retrofit projects, the installed new systems may have to temporarily interact with the existing legacy systems.
The bottom line
The argument for a single lifecycle partner is not about convenience; it is about where risk and cost accumulate. When one team supports the build, the maintenance, and the upgrade, the institutional knowledge of the plant never leaves, the documentation is continuous, and multi-OEM coordination stays with a vendor-neutral integrator who can select best-fit components and still support them through warranty and operations — leaving the operator backed by one accountable party rather than a rotating set of vendors whose records and incentives do not align.
Critical power is not a one-time install. Its value compounds — or erodes — over decades, in the gaps between the build, the maintenance, and the upgrade. The build sets a baseline designed to be maintained; the maintenance protects that baseline and feeds the upgrade; the upgrade re-establishes it for the next cycle. Run as three disconnected procurements, that loop breaks at every handoff. Run as one continuum — supported by a single accountable systems integrator with nationwide in-house field-service teams — it holds, which is the most reliable way to lower operational risk and total cost of ownership over the life of a facility.