What I’ve Seen Over 20 Years: The 5 Most Common Mistakes in Power Quality

I have been working in the field of power quality for over twenty years. During this time, we have taken measurements at hundreds of facilities, commissioned hundreds of compensation and filtering systems, and re-evaluated dozens of them in response to the question, “Why isn’t it working?”

I have been working in the field of power quality for over twenty years. During this time, we have taken measurements at hundreds of facilities, commissioned hundreds of compensation and filtering systems, and re-evaluated dozens of them in response to the question, “Why isn’t it working?” Every industry has its own unique load profile: textiles, steel, plastics, food, hospitals, data centers… But here’s the interesting part: Even though the facilities differ, the mistakes made are almost always the same.

An investment in power quality is not merely an expense incurred to avoid reactive power penalties. When properly designed, it extends equipment lifespan, ensures production continuity, and reduces transformer and cable losses. When poorly designed, however, it generates operational costs several times the initial investment. Just because the bill says “penalty eliminated” doesn’t mean the system is healthy; sometimes the penalty is gone, but the capacitors keep overheating, the drives keep tripping, and the neutral conductor keeps swelling.

In this article, I want to discuss the five most common mistakes I see in the field. No names, no brands—just realistic scenarios, costs, and the right approach. Some of these mistakes stem from the technical team, some from the procurement process, and some from the “it’ll do for now” reflex. The common thread is this: They are all preventable.

My goal is not to make a sale. My goal is to ensure that decision-makers and engineers do not fall into the same traps again.

Mistake 1: The “We Have Power Factor Correction, That’s Enough” Fallacy

One of the phrases I hear most often in the field is: “We already have power factor correction.” When you open the panel cover, there’s a capacitor bank, the relay is operating, and the power factor (cosφ) target is set to 0.98. On paper, everything looks fine. But inside the facility, there are frequency converters, UPSs, LED drivers, and rectifiers—in other words, plenty of harmonic sources. What does a reactor-less, purely capacitive compensation system do in this environment? It goes into resonance.

Real-Life Case Study

At a plastic injection molding facility that heavily utilized VFDs, a reactorless capacitor bank had been put into service. Within a few months, capacitors began to fail one after another, contactors stuck, and fuses started blowing. Facility management initially assumed the “materials were substandard”; however, the problem lay not in the materials but in the system architecture. Measurements revealed that the dominant 5th and 7th harmonics were resonating at the compensation system’s resonant frequency. The solution was to switch to a detuned reactor-based compensation system and resize it according to the load profile. The costs incurred up to that point—including spare parts, labor, and lost production—exceeded the initial investment difference for the correct system.

Cost and Risk

Capacitors operating in a harmonic environment are exposed to excessive current. Premature failure, transformer overheating, cable losses, and unplanned outages follow one after another. When power quality is poor, the lifespan of electrical equipment is shortened; this is a cost that doesn’t appear on the bill but accumulates in the maintenance budget. Moreover, the false sense of security that “compensation is in place” prevents the real problem—harmonic distortion—from being detected for a long time.

The Right Approach

Before installing compensation, measure the facility’s load characteristics and harmonic profile. If harmonic distortion is high, consider compensation with reactor-based (detuned) filters or active solutions instead of a pure capacitive system. If you have an existing bank and are experiencing frequent capacitor/reactor failures, this is often a sign that the “system is operating at the wrong frequency,” not that the “materials are defective.” Saying “we have compensation” does not mean “there are no power quality issues.” Compensation is a tool; it works when the right tool is placed in the right place.

Mistake 2: Purchasing Equipment Without Performing a Harmonic Analysis

The second major mistake is buying without measuring. People look at the bill, see a reactive power penalty, and say, “Let’s install a filter.” THDi, THDv, dominant harmonic orders, Isc/IL ratio… None of these are known. Then, either a device with insufficient capacity is purchased, or the wrong technology is chosen. AHF, SVG, or a passive filter? The answer to this question cannot be determined without measurements.

Real-Life Case

At a textile factory, the project proceeded based on the assumption that “there are harmonics” without conducting comprehensive measurements prior to installation. During the commissioning process, the true picture emerged: approximately 1,030 A of line current per phase, with a THD value exceeding 68 percent. This was a far more severe level of distortion than originally assumed. After a four-module active harmonic filter (AHF) with a capacity of 600 A was commissioned, the THD dropped to 2 percent. Fortunately, the system was modular and scalable; otherwise, a single incorrectly sized device would have rendered the investment useless. The key lesson was this: If measurements had been taken, the capacity and architecture would have been designed correctly from the start; commissioning would have resulted in a planned outcome rather than surprises.

Cost and Risk

If harmonic analysis is skipped, the selected filter will either be inadequate or cause damage to the system. IEEE 519 and EN 50160 define limits for harmonic currents injected into the grid and voltage quality. Exceeding these limits carries the risk of penalties on the distribution side. An incorrect choice, on the other hand, results in wasted investment: A device initially purchased as “cheap” is later replaced with the “correct” one; the difference lies in lost production and rework.

The Correct Approach

Conduct at least one week of continuous measurements using an IEC 61000-4-30 Class A-compliant power quality analyzer. A single-shift “snapshot” is insufficient; the production cycle, day-night differences, and weekend profile must be visible. The measurement point should preferably be the point of common connection (PCC). Do not request a quote until THDi, THDv, dominant harmonics, and the load variation profile are clearly established. Whether AHF, SVG, or a hybrid solution is required can only be determined with this data. “Installing without measuring” is the most costly habit in power quality.

Mistake 3: Choosing the Cheapest Bid

At the procurement table, the most attractive figure is usually the lowest unit price. The technical specifications appear “adequate,” a warranty is said to be “included,” and delivery is “fast.” A decision is made. Two years later, the equipment breaks down frequently, the manufacturer cannot provide technical support, and replacement parts are delayed. The replacement cost exceeds the initial investment; production losses are a separate expense. The cheapest bid is often the most expensive over the life cycle.

Real-Life Case

A low-cost AHF solution was selected at a medium-sized industrial facility. In the first year, it was deemed “acceptable.” In the second year, breakdowns began one after another. The fan, control board, communication module… A different part each time. The supplier said, “We’re waiting for the manufacturer”; delays stretched out over weeks. The facility was eventually forced to replace the system. The replacement cost was several times the initial “savings.” When the cost of unplanned downtime was added, the life-cycle calculation became clear: The cheap bid had turned out to be an expensive lesson.

Cost and Risk

Life-cycle cost (LCC) includes purchase, installation, commissioning, maintenance, spare parts, the device’s own losses, and potential downtime. It is well known that the focus on the “lowest bid” in public tenders in Turkey leads to higher operational costs in the long term; this is precisely why the “most economically advantageous bid” approach exists. Efficiency is also a component of LCC. For example, SiC-based power electronics can significantly reduce losses compared to conventional IGBTs; this difference translates into savings on energy bills over the years. A non-standard or poorly supported product, however, also poses risks in terms of compliance with IEC 61000, EN 50160, and IEEE 519 standards.

The Right Approach

Don’t evaluate a proposal based solely on the price column. Request a 10-year YDM comparison. Ensure that efficiency, warranty coverage, SLA, spare part lead time, firmware access, and commissioning reports are specified in writing. “Cheapest” is not the same as “most advantageous.” A cheap mistake in power quality turns into an expensive failure.

Mistake 4: Neglecting the Maintenance Contract

The system has been commissioned. The report has been signed. Everyone is happy. Then… no one checks on it. The “we’ll look into it when a failure occurs” approach leads to the same result for both active power electronics devices and passive systems: unplanned downtime.

Active filters and SVGs may not have frequently failing parts like mechanical contactors; this does not mean they can be left without maintenance. Fans get dusty, filters clog, connection points overheat, firmware isn’t updated, and alarms aren’t tested. One day, thermal protection kicks in and the system shuts down—“usually at the worst possible time.”

Real-Life Case

In an AHF system without a maintenance plan, the cooling fans had become clogged over time. The internal temperature exceeded the critical threshold, thermal protection kicked in, and filtration stopped. The facility first noticed the resurgence of harmonics due to equipment overheating, followed by complaints about reactive power and power quality. Procuring a replacement fan, cleaning, inspection, and restarting the system took several days. The production loss during those few days was many times greater than the cost of annual periodic maintenance. Warranty conditions were also contingent on “regular maintenance”; this neglect had jeopardized not only operations but also the company’s rights.

Cost and Risk

Neglected maintenance leads to fan failure, IGBT or capacitor damage, unplanned downtime, and emergency repairs. Emergency repairs are always more expensive than scheduled maintenance. Furthermore, without monitoring, pre-failure warnings are missed, and the problem escalates. A maintenance contract is not an “additional cost” but rather insurance for your investment.

The Right Approach

Include the maintenance schedule in the contract at the time of commissioning. The scope should include at least the following: fan and filter cleaning, connection checks using thermal imaging, firmware updates, protection and alarm tests, and verification of DC bus and cooling performance. Set up remote monitoring (Modbus/Ethernet); review the data periodically. “We’ll look into it when a failure occurs” is a luxury we can’t afford when it comes to power quality—and it usually ends up costing a lot.

Mistake 5: Dealing with the dealer instead of the manufacturer

Let me be clear here: The problem is not the “dealer” model itself. A qualified business partner who has direct access to the manufacturer and can manage testing and commissioning is extremely valuable. The problem lies in making a selection based solely on price and delivery time, without verifying these capabilities. An intermediary that cannot access the firmware, cannot escalate fault analysis to the manufacturer, or cannot stock spare parts will leave you stranded in a crisis.

Real-Life Case

A facility had compensation and filtering equipment from various brands. Every time a failure occurred, a different technical team would arrive, and each team would say, “It’s not our device; it’s the other side’s.” There was no integrated monitoring; the source of the failure remained unclear. Downtime was increasing, and the production schedule was being disrupted. The problem wasn’t a single faulty part—it was fragmented accountability. When the facility switched to manufacturer-led engineering and centralized technical support, both diagnosis and response times became predictable. There were “a bunch of devices” on the same panel; what was missing was a single engineering team to take full responsibility for them.

Cost and Risk

Firmware updates, parameter optimization, and in-depth fault analysis are often performed at the manufacturer level. Without this access, the response time becomes uncertain. Warranty coverage also becomes unclear: Was the installation performed in accordance with the manufacturer’s procedures, or not? If spare parts cannot be obtained, the system is effectively unreliable even if it appears to be “operational.” Fragmented supply means fragmented responsibility; in a crisis, when everyone is supposedly responsible, no one actually is.

The Right Approach

Evaluate the supplier along four key dimensions:

1. Measurement Competence — Can they provide IEC 61000-4-30 Class A measurements and a report prior to installation?
2. Manufacturer Technical Access — Is there direct access to the manufacturer for firmware, parameter, and fault analysis?
3. Engineering Accountability — Is the commissioning report signed, and is the technical responsibility for the project clearly defined?
4. Spare Parts and SLA — Are lead times and warranties for critical components documented in writing?

If you’re working with a distributor, ensure these four points are secured in writing. If you’re working directly with the manufacturer, ask the same questions—because it’s not about the title, but about competence and access. In energy management, the engineering and support structure behind the product are just as critical to the investment as the product itself.

Conclusion: The difference lies not in recognizing the mistake, but in not repeating it

These five mistakes are patterns I’ve seen time and again in the field over the past twenty years:

1. Assuming that power quality is fully addressed simply because compensation is in place
2. Purchasing equipment without performing a harmonic analysis
3. Assuming the cheapest bid is the most advantageous one
4. Putting off the maintenance contract with a “we’ll deal with it later” attitude
5. Making decisions without questioning the manufacturer’s engineering and technical access

The common thread among all of these is this: Short-term convenience leads to long-term costs. Measurements aren’t taken, life cycles aren’t calculated, responsibility is diffused, maintenance is postponed, and the system eventually comes back with the question, “Why isn’t it working?”

At Aha Technology, our approach is to make the opposite of these mistakes the standard. As a local manufacturer, we tailor AHF, SVG, STATCOM, and compensation solutions to the actual problems on-site. With SiC technology and a modular architecture, we prioritize both efficiency and scalability. Pre-installation measurements, the right technology selection, signed commissioning, and continuous monitoring and maintenance are not “extras” for us after the sale; they are the solution itself. Since we design the product ourselves, the path to firmware, parameters, and fault analysis is short; field feedback is directly reflected in the next product.

My advice is clear: For your next energy quality decision, measure first, then choose; focus on the life cycle, not just the price; include maintenance in the contract; and ask about your counterpart’s technical access and engineering responsibility. Facilities that ask these questions avoid most of the five mistakes I’ve seen over the past twenty years. Those who don’t ask, however, generally end up reliving the same story—just on a different panel.

Because when it comes to power quality, the real cost isn’t the device’s list price. The real cost is the time lost due to a wrong decision—and that time is gone forever.

Serkan ANBAR
CEO, Aha Teknoloji