Electric Vehicle Chargers and Residential Power Quality — A Two-Week Field Study
| Site | Residential condominium — Bangkok, Thailand |
| Measurement duration | Two weeks of continuous field measurement |
| EV charger type | Single-phase Level 2 AC chargers (dominant usage pattern) |
| Voltage unbalance | Average 0.535% · Peak 2.18% during active charging |
| Current THD during charging | 15–20% — significant increase over baseline |
| Neutral current | Significant — driven by single-phase loading dominance |
| Infrastructure condition | Retrofitted — not originally designed for EV charging loads |
| Key finding | Single-phase EV charging creates phase imbalance and harmonic injection that existing residential wiring was not designed to handle |
01 Context and Background
Electric vehicle adoption is accelerating globally, and with it comes a set of power quality challenges that most residential distribution infrastructure was never designed to handle. This field study by Ngaopitakkul (2025) at Chulalongkorn University examines the PQ impact of EV charger deployment in a real residential condominium in Bangkok, Thailand — a city where rapid EV adoption is outpacing electrical infrastructure upgrades.[1]
The study is particularly relevant because it addresses the most common real-world scenario: retrofitted residential infrastructure with single-phase Level 2 AC chargers installed in individual parking spaces or garages, without coordinated load management. This is not a purpose-built EV facility with balanced three-phase charging and smart load control — it is the existing electrical system of a multi-unit residential building confronted with a new generation of high-power non-linear loads that were not anticipated in the original design.
The challenge EV chargers present to distribution networks is different in character from the industrial harmonic problems that dominate most PQ literature. Industrial harmonic sources are concentrated, predictable, and often subject to utility connection studies. Residential EV chargers are distributed, stochastic, and connected to the low-voltage network at points that were never designed to absorb significant harmonic or unbalanced loads. A utility engineer planning distribution upgrades needs field data from exactly this type of setting.
02 Measurement Methodology
A two-week continuous field measurement campaign was conducted at the condominium’s electrical distribution panel. The measurement duration is important: EV charging behaviour varies with time of day, day of week, and individual resident patterns. A two-week dataset captures the variability that a single-day or single-event measurement would miss, and provides a statistically meaningful characterisation of the PQ impact under realistic usage conditions.[1]
The following parameters were measured and analysed:
- Load profiles — current demand versus time, revealing charging patterns
- Current and voltage waveforms — three-phase RMS values and waveform capture during charging events
- Phase symmetry — distribution of charging load across the three phases
- Harmonic distortion — current THD analysis during active charging and at baseline
- Voltage unbalance — quantified using the maximum deviation method (NEMA definition): maximum deviation of any phase voltage from the three-phase mean, divided by the mean
- Neutral current — magnitude and waveform during charging events
This study used the NEMA maximum deviation method for voltage unbalance, which requires only phase voltage magnitudes. The IEC symmetrical components method (ratio of negative-sequence to positive-sequence voltage) is more rigorous but requires phasor measurements. For small unbalances below approximately 3%, both methods give similar numerical results. The peak unbalance of 2.18% reported in this study would be close to the EN 50160 limit of 2% under either definition — making the finding relevant to both North American and IEC practice.[2]
03 Key Findings
Measured PQ parameters
| Parameter | Baseline (no charging) | During active EV charging | Standard limit | Status |
|---|---|---|---|---|
| Current THD | Low — normal residential baseline | 15–20% | IEEE 519: 5–8% TDD at LV PCC | ELEVATED |
| Voltage unbalance (avg) | Below 0.5% | 0.535% average | EN 50160: ≤ 2% (95% of week) | Within limit |
| Voltage unbalance (peak) | Below 0.5% | 2.18% peak | EN 50160: ≤ 2% (95% of week) | AT LIMIT |
| Neutral current | Low | Significantly elevated | Not directly limited — sizing risk | MONITOR |
| Phase loading symmetry | Approximately balanced | Single-phase dominant | Balanced loading assumed in design | IMBALANCED |
| Source: Ngaopitakkul (2025).[1] Two-week field measurement, Bangkok condominium. Voltage unbalance by NEMA maximum deviation method. | ||||
Single-phase charging dominates — the root of the unbalance problem
The most significant behavioural finding was that single-phase charging completely dominated the observed usage patterns. Residents used single-phase Level 2 AC chargers, connected to whichever phase their parking space happened to be served by — without any coordination across phases. The result is a highly unbalanced load that varies with how many residents charge simultaneously and which phase they are on.[1]
This is not a technology limitation of EV chargers — three-phase chargers exist and would distribute the load symmetrically. It is a consequence of how residential EV charging is deployed in practice: individual units, individual chargers, no system-level coordination, connected to whatever single-phase branch is available at the parking level.
The condominium’s electrical infrastructure was not originally designed for EV charging loads. In retrofitted buildings, the effects of EV charging are often exacerbated by design limitations: neutral conductors sized for the pre-EV load, distribution panels without spare capacity, and wiring runs that were not engineered for the combination of high current magnitude and significant harmonic content. A building designed today with EV charging in mind would have a fundamentally different electrical architecture — three-phase chargers, managed charging systems, oversized neutrals, and a distribution capacity assessment that includes the EV load from the outset.
04 Technical Analysis
EV chargers as non-linear loads
EV on-board chargers use switched-mode power conversion — the AC supply is rectified and then regulated by a high-frequency switching converter to deliver the required DC current to the battery. This makes the EV charger a non-linear load that draws current in pulses rather than a sinusoid, injecting harmonic currents into the supply network. The dominant harmonic orders depend on the converter topology, but third, fifth, and seventh harmonics are typical for single-phase chargers.[1]
The current THD values of 15–20% measured during charging are consistent with published data for residential Level 2 AC chargers without active power factor correction. Modern chargers with active front-end PFC circuits can achieve THD below 5%, but these are not uniformly deployed in the existing residential EV charging population.
Phase imbalance — the neutral current consequence
When single-phase loads dominate on one or two phases, the three-phase system becomes unbalanced. In a four-wire system, the neutral conductor carries the vector sum of the three phase currents — which under balanced sinusoidal conditions is zero. Under the unbalanced single-phase EV charging conditions observed in this study, the neutral current became significant, creating thermal loading on a conductor that was sized for a much smaller current. This is the identical mechanism to the triplen harmonic neutral overloading discussed in hospital PQ case studies — different root cause, same consequence for the neutral conductor.
Voltage unbalance — peak at EN 50160 limit
The average voltage unbalance of 0.535% is well within the EN 50160 limit of 2%. However, the peak of 2.18% during simultaneous charging events on the same phase approaches and momentarily exceeds the limit. For the utility engineer, this is important: EN 50160 compliance is assessed as a 95th-percentile statistic over a one-week observation period — a single peak event during one evening does not by itself constitute non-compliance. But it signals that as EV penetration increases in the building, peak unbalance will increase proportionally, and the statistical distribution will shift toward the limit.
A single condominium with a dozen EV chargers produces a peak unbalance near the EN 50160 limit. A distribution feeder serving ten such buildings — which is a realistic near-term scenario in any urban area with active EV adoption — could produce sustained unbalance that exceeds the limit at the LV feeder level without any individual building being in violation of its connection conditions. This is a network-level PQ planning problem that requires monitoring at the feeder level, not only at individual customer connections.
05 Recommendations
The study identifies the following engineering measures for EV-ready residential systems:[1]
- Three-phase EV chargers — distribute the charging current symmetrically across all three phases, eliminating the single-phase imbalance problem at source. Appropriate for new installations and major retrofits
- Coordinated load management (smart charging) — control charging schedules to avoid simultaneous peak demand on the same phase, reduce peak current, and allow time-of-use management. Requires a building energy management system and charger communication capability
- Phase assignment planning — for single-phase charger installations, deliberately assign chargers to alternate phases across the building to balance the load. Requires inventory of phase assignments at all parking spaces
- Neutral conductor assessment — review and upsize neutral conductors where single-phase EV charging loads are concentrated. The neutral conductor in a pre-EV building was sized for a balanced load assumption that no longer holds
- Active harmonic filtering — if the charger population cannot be replaced with PFC-equipped units, a central active harmonic filter at the distribution panel can reduce THD to acceptable levels
- Distribution capacity study before deployment — any residential building adding EV charging should conduct a load flow and PQ impact study before installing chargers, not after problems appear
Before any capital expenditure, the most immediate and zero-cost mitigation is phase assignment planning. Auditing which parking spaces are served by which phase and deliberately assigning new EV charger connections to the least-loaded phase costs nothing and directly reduces the peak unbalance. It will not solve the harmonic problem, but it addresses the dominant cause of neutral current overloading and voltage unbalance at no capital cost.
06 Power Quality Perspective
This case study represents a class of PQ problem that will only grow in frequency and severity over the next decade: the retrofit of new high-power non-linear loads onto distribution infrastructure that was designed for a completely different load profile. EV chargers, heat pumps, and battery storage systems are all being connected to residential and commercial buildings whose electrical systems were designed for incandescent lighting, resistive heating, and linear motor loads.
From a utility distribution planning perspective, the most important finding in this study is not the individual building numbers — 15–20% THD, 2.18% peak unbalance. These are manageable at a single-building scale. The planning concern is the aggregate: when a significant fraction of homes and condominiums on the same LV feeder deploy single-phase EV chargers with uncoordinated charging, the feeder-level unbalance and harmonic distortion can reach levels that affect all customers on the feeder, not only the EV charging buildings.
Utility engineers who spent their careers managing harmonics from industrial loads and capacitor switching transients from distribution feeders are now facing a distributed, residential-scale version of the same problem — amplified by the speed of EV adoption and the absence of utility-visible metering at the individual charger level. The traditional utility PQ toolbox — PCC measurements, IEEE 519 compliance studies, harmonic filter specifications for large industrial customers — was not designed for millions of 7 kW residential loads distributed across the LV network. New monitoring strategies, new planning tools, and new connection conditions for residential EV charging are all in active development. This field study from Bangkok is one data point in a dataset that the industry urgently needs to build.
References
- Ngaopitakkul A. “Evaluating Effects of Electric Vehicle Chargers on Residential Power Infrastructure.” Applied Sciences, vol. 15, no. 11, p. 5997, 2025. DOI: 10.3390/app15115997. Open access under CC BY 3.0.
- EN 50160:2010+A3:2019. Voltage characteristics of electricity supplied by public electricity networks. CENELEC, Brussels.
- IEEE Std 519-2022. IEEE Standard for Harmonic Control in Electric Power Systems. IEEE, New York, NY, 2022.
- NEMA MG-1-2021. Motors and Generators — Voltage Unbalance Definition. National Electrical Manufacturers Association, Rosslyn, VA.
Ngaopitakkul A. “Evaluating Effects of Electric Vehicle Chargers on Residential Power Infrastructure.” Applied Sciences, 15(11), 5997, 2025.
DOI: 10.3390/app15115997 · Read the original article at MDPI →
Published open access under CC BY 3.0. This case study is presented in summary and commentary form. The PQ Perspective section (Section 6) is original IPQDF editorial commentary by Denis Ruest, M.Sc. (Applied), P.Eng. (ret.). IPQDF does not claim authorship of the original research.