Dirty Power, Safe Buildings: How Power Quality Silently Drives Failures

Why power quality problems get misdiagnosed

This episode of Built, Wired and Secured starts with a familiar operational mystery: servers rebooting during the busiest hour while every sign seems to point to software. The turning point comes when someone notices the hallway lights dipping every time a rooftop unit cycles. That is the core lesson of the conversation between Alex Morgan and Michael Harrington: many incidents that look like application, network, or device failures are actually power quality problems hiding in plain sight.

The discussion makes the case that “dirty power” is not an abstract electrical engineering topic. It is an operational risk that shortens equipment life, triggers false alarms, creates intermittent outages, and wastes time by sending teams in the wrong direction. Instead of framing the issue as a technical curiosity, the episode keeps returning to a practical question: what breaks if this goes down? That question helps facilities, IT, and ownership teams decide where to focus attention first.

What “dirty power” means in everyday operations

Michael defines dirty power in practical terms rather than academic ones. The conversation points to a group of issues that often do not make it onto formal incident boards but still make equipment unstable:

• Voltage sags
• Spikes
• Harmonics
• Grounding oddities

The important point is not just the definition. It is how these problems show up in the field. According to the episode, operators should pay attention when they see:

• Intermittent equipment reboots
• Alarms with no clear cause
• Lights that dim or flicker for a blink
• Error logs with no recent updates
• Nuisance breaker trips
• Events that correlate with time of day or mechanical cycles

One of the most useful takeaways is that different teams notice different clues. Tenants may complain that an application froze. IT may focus on logs and reboots. Facilities may be the only group that notices a half-second lighting dip when a large motor starts. The episode argues that those clues need to be combined, not handled in isolation.

Common causes in campuses and multi-tenant properties

The discussion avoids overcomplicating root cause analysis, but it does outline the usual suspects property and operations teams should keep in mind. Michael highlights several recurring causes on campuses and multi-tenant sites:

• Shared tenant loads
• Large motors or welders cycling on and off
• Aging or poorly maintained UPS systems behaving badly under partial loads
• Bad grounding and bonding that create stray voltages and communications noise
• Utility or transformer interactions involving resonance or voltage regulation

Another strong point from the episode is that there is rarely one single villain. The real issue is often a combination of conditions. That matters because it changes the response. Teams should not jump straight to the most expensive fix. They should prioritize based on impact and evidence.

Why monitoring comes before major spending

One of the clearest frameworks in the episode is the order of operations for response and investment. Alex lays out a sequence: monitoring first, conditioning second, UPS strategy after that, and major capital decisions like microgrids only when justified. Michael agrees with that structure and explains why.

Monitoring matters because it creates timestamps that can be correlated with IT logs, alarm systems, and operational complaints. Without that data, teams are often replacing hardware, blaming software, or escalating vendors without proving what happened. In other words, monitoring converts vague suspicion into a pattern that can be tested.

The episode also emphasizes that selective conditioning can often solve the real issue at a much lower cost than broad system replacement. The message is not to underinvest. It is to match the fix to the failure mode.

The UPS myth: why bigger is not always better

One of the strongest operational warnings in the episode is about UPS strategy. Michael pushes back on the assumption that more UPS capacity automatically means more resilience. Oversizing, he explains, can actually mask battery health problems. If a UPS runs at low load, transfer logic can introduce momentary blips when switching sources.

That becomes especially important because the failures that matter most are often very short. A half-second event may never stand out in paperwork, but it can still reset an access controller or disrupt a server power supply.

The practical advice is to stop evaluating UPS equipment by capacity alone. Teams should:

• Test transfer curves under real load
• Verify actual battery runtime
• Watch for micro interruptions during switching
• Evaluate whether a smaller targeted adjustment would solve the issue

The episode even suggests that modest changes, such as adding a few seconds of ride-through or placing a line conditioner on a rack, may remove the pain point for a fraction of the cost of a bigger UPS project.

Three real-world examples that make the issue concrete

The conversation includes three concise examples that show how easily power quality issues can be mistaken for unrelated failures.

• Clinic EHR freezes: The symptom was intermittent application hangs. The root cause was a lab centrifuge on the same panel creating voltage sags. The immediate fix was a dedicated conditioned circuit for the clinic plus basic power logging.
• Repeated false fire alarms on a campus: The symptom was detectors tripping at odd hours. The root cause was harmonic distortion from VFDs in the mechanical room. The fix was to add harmonic filters and rebalance loads.
• Nightly network closet outages in an office tower: The symptom was evening connectivity drops. The root cause was utility voltage regulation drift during peak periods. The fix involved installing power quality meters, negotiating staged utility fixes, and adding local conditioning on the most sensitive floors.

These examples reinforce the show’s central point: if the symptom looks random, do not assume the cause is random. Look for operational patterns and electrical triggers.

Why microgrids are not the first answer

The episode deliberately cools off the temptation to jump to high-visibility capital projects. Michael makes the point clearly: microgrids may be powerful, but they are not the right answer for intermittent sags caused by something like a tenant motor. They also introduce their own overhead, including fuel logistics, controls, and maintenance requirements.

That does not mean distributed resiliency has no place. It means teams should fix immediate power quality issues first, then move to long-duration autonomy solutions only when the business case supports them.

The quarter-by-quarter checklist listeners can use

The episode closes with a practical checklist that property, facilities, and IT leaders can use right away:

• Enable timestamped power logging on panels serving critical loads
• Correlate those timestamps with IT or alarm logs before replacing equipment
• Visually inspect grounding and bonding during maintenance windows
• Identify large single-source loads on shared panels and consider dedicated circuits
• Review UPS age, runtime, and transfer behavior, then test under load
• Add targeted conditioning for chronic trouble spots before replacing major equipment
• Only after that, evaluate generators or microgrids if the business case requires long autonomy

There is also a useful communication point for vendor management. When bringing in an electrical contractor, the team should provide correlated event logs and describe the operational impact. That changes the conversation from a vague concern to a testable pattern.

Final takeaway

This episode is a strong reminder that resilience is not only about backup power or bigger hardware. It is about seeing weak signals early, connecting facilities clues with IT symptoms, and making decisions in the right order. Monitor first. Correlate evidence. Apply targeted fixes. Spend big only when the impact justifies it. For buildings and campuses that rely on stable digital operations, that approach can reduce downtime, prevent unnecessary replacement cycles, and extend the life of critical systems.

Power quality is an operations problem long before it becomes an electrical project

Many building and operations teams have lived through some version of the same scenario: a critical system starts failing during busy hours, logs point in one direction, vendors point in another, and everyone assumes the issue is software, networking, or a bad device. Then someone notices a flicker in the lights at exactly the same time a mechanical load kicks on. Suddenly the problem looks very different.

That is the central message of this episode of Built, Wired and Secured, where Alex Morgan talks with Michael Harrington about how power quality quietly creates failures that get misdiagnosed. The conversation keeps the topic grounded in building operations rather than electrical theory. Instead of treating power quality as a niche engineering concern, the episode frames it for facilities leaders, IT teams, and property operators who need fewer unexplained incidents, more reliable systems, and smarter capital decisions.

The key insight is simple: some of the most frustrating outages are not caused by the application, the switch, or the controller itself. They are caused by the quality of the power feeding those systems.

What dirty power looks like in the real world

In the episode, dirty power refers to the kinds of electrical conditions that often stay off the incident board even though they cause real pain in the field. That includes voltage sags, spikes, harmonics, and grounding oddities. These issues are easy to overlook because they may last only a fraction of a second, but that can be enough to reset sensitive electronics, trigger nuisance alarms, or shorten equipment life over time.

What makes the discussion useful is the way it translates those ideas into operational symptoms. Teams should pay attention when they see:

• Intermittent equipment reboots
• Application hangs with no recent updates or code changes
• Alarm activations that do not point to a clear cause
• Breaker trips that feel inconsistent or hard to reproduce
• Lights that dim or flicker briefly
• Failures that happen at a specific time of day or during known mechanical cycles

That last point matters. The episode emphasizes patterns over isolated events. If a server power supply trips every time a large motor starts, or if access control devices reset when another building system cycles, the problem may never be obvious from IT logs alone. Facilities may notice lighting dips. Tenants may notice application slowdowns or freezes. Operators need both views to connect the dots.

Why so many teams chase the wrong root cause

One reason power quality problems persist is that the visible symptom often appears several layers away from the source. A failing application gets escalated to software support. Connectivity drops get blamed on network hardware. False alarms get treated as panel or detector issues. But in the examples shared in the episode, the underlying trigger was electrical.

That matters because misdiagnosis wastes money in two ways. First, teams spend time replacing or troubleshooting the wrong equipment. Second, they often skip the evidence-gathering step and jump straight to expensive fixes that do not address the real failure mode.

Michael brings the conversation back to a practical decision lens: what breaks if this goes down? That question helps separate critical loads from acceptable short-term disruptions. It also helps organizations decide whether the right next move is monitoring, conditioning, UPS work, or a larger resiliency strategy.

The usual suspects on campuses and multi-tenant sites

The episode outlines several common causes of dirty power in campuses and commercial properties. None of them are exotic. In fact, that is part of the warning. These are normal conditions that create abnormal outcomes when they affect sensitive systems.

• Shared tenant loads can create instability when different kinds of equipment compete on the same electrical infrastructure.
• Large motors or welders cycling on and off can create sags that are short in duration but significant in effect.
• Aging or poorly maintained UPS systems may behave unpredictably, especially under partial loads.
• Grounding and bonding problems can create stray voltage and communications noise.
• Utility or transformer interactions can introduce resonance or voltage regulation issues that show up at specific times.

The discussion also avoids the trap of searching for one perfect villain. In many cases, the problem is a combination of factors. That is why pattern recognition and measured testing matter more than assumptions.

Start with monitoring, not guesswork

One of the best frameworks in the episode is the order in which leaders should approach these issues. Monitoring comes first. That may sound unexciting compared with major capital projects, but it is the step that turns suspicion into evidence.

Timestamped power logging on panels serving critical loads gives teams something they can correlate with other records. If a reboot, alarm, or nightly outage lines up with a measured electrical event, the conversation changes immediately. Instead of saying, “we think something is wrong,” teams can say, “this event occurred at this time, under this condition, and here is the operational impact.”

That kind of correlation is useful internally, but it is also critical when bringing in outside contractors. The episode makes a strong point about handoffs: give vendors the event patterns and the operational consequences. That changes the engagement from broad troubleshooting to scenario-based testing.

Selective fixes often beat expensive blanket solutions

Another practical theme in the episode is that fixes should be prioritized, not reflexive. Once teams understand the impact and have some evidence, they can pursue more surgical interventions.

Those targeted fixes might include:

• Dedicated circuits for especially sensitive loads
• Basic power logging in known problem areas
• Line conditioning at a rack or closet
• Harmonic filters in areas affected by VFDs
• Load rebalancing on shared panels

The value of this approach is business discipline. Not every dirty power issue requires a large redesign. In many cases, a focused intervention can reduce outages, protect equipment, and limit operational disruption without forcing a full electrical overhaul.

The important warning about UPS strategy

The episode also challenges a common resilience assumption: that a bigger UPS is always a better UPS. Michael explains that oversizing can hide battery health problems. If a UPS runs at low load, transfer behavior may create momentary blips when switching sources. Those blips may be brief enough to avoid standing out in documentation but still long enough to break devices in the real world.

That is why teams should not evaluate UPS investments on capacity alone. The better questions are operational:

• How old is the UPS?
• What runtime does it actually deliver?
• How does it transfer under real load?
• Are there micro interruptions during switching?

In some cases, the right answer may be surprisingly modest. A few seconds of additional ride-through or a targeted line conditioner may solve the actual pain point for a fraction of the cost of a major UPS expansion.

Three examples that show how hidden the issue can be

The episode includes three anonymized examples that make the risk very concrete.

In a clinic, an EHR system experienced intermittent application hangs. The root cause was not software at all. A lab centrifuge on the same panel was creating sags. The immediate fix was a dedicated conditioned circuit and basic power logging.

On a campus, teams dealt with repeated false fire alarms at odd hours. The cause turned out to be harmonic distortion from VFDs in the mechanical room. Harmonic filters and load rebalancing addressed the issue.

In an office tower, network closets were losing connectivity nightly. The source was utility voltage regulation drift during peak periods. The response involved power quality meters, staged utility coordination, and local conditioning on the most sensitive floors.

These examples are especially useful because they show different types of business impact: clinical operations, life-safety nuisance events, and tenant-facing connectivity problems. In every case, the first visible symptom could have sent the team down the wrong path.

Why microgrids come later, not first

The episode does not dismiss major resiliency infrastructure, but it does put it in order. Microgrids may be powerful tools for long autonomy, yet they are not a cure for intermittent sags caused by shared loads or tenant equipment. They also introduce meaningful overhead in the form of controls, maintenance, and fuel logistics.

That is an important operational mindset. Large capital solutions should solve a defined business problem, not serve as a substitute for diagnosis. If the real issue is power quality at a particular load, fix that first. If the organization later needs autonomous operation for critical systems, then evaluate generators or microgrids through a clear business case.

A practical checklist for this quarter

The episode closes with a strong action sequence that teams can use immediately:

• Enable timestamped power logging on panels serving critical loads.
• Correlate those timestamps with IT or alarm logs before replacing equipment.
• Inspect grounding and bonding during maintenance windows.
• Identify large single-source loads on shared panels and consider dedicated circuits.
• Review UPS age, runtime, and transfer behavior, then test under load.
• Add targeted conditioning for chronic trouble spots before replacing major equipment.
• If long autonomy is truly required, evaluate generators or microgrids with a business case approach.

That order matters because it keeps cost proportional to impact. It also keeps teams from escalating too quickly to the most visible or expensive answer.

The business lesson behind the electrical lesson

At its core, this episode is not only about voltage quality. It is about operational discipline. Random-looking failures are often pattern-driven. The organizations that recover fastest and spend smartest are the ones that gather evidence, correlate symptoms across teams, and target the root cause before overinvesting.

For commercial real estate operators, facilities teams, and IT leaders, that translates into fewer unexplained incidents, better vendor conversations, more credible capital planning, and longer life from critical equipment. If your environment includes sensitive systems, shared infrastructure, and tenant expectations for uptime, power quality deserves a place in your risk and maintenance conversations.

If this topic hits close to home, listen to the full episode for the complete discussion and the step-by-step checklist. It is a practical conversation that helps turn an often-missed technical issue into a clear operations strategy.